[Request] Parsed ggml-turbo/request0.json [Request] seed=42 lm_batch=1 synth_batch=1 [Request] caption: An upbeat and anthemic pop-rock track driven by bright, slig... [Request] lyrics: 490 bytes [Request] bpm=83 dur=88 key=G major ts=4 lang=fr [Request] lm: temp=0.85 cfg=2.0 top_p=0.90 top_k=0 [Request] dit: steps=8 guidance=0.0 shift=3.0 [Request] task_type: text2music [Request] solver: euler (stork_substeps=10) [Request] lm_mode: generate [Request] output_format: mp3 [Request] synth_model: acestep-v15-turbo-Q5_K_M.gguf [Request] vae: vae-BF16.gguf [Request] audio_codes: (present) [Ace-Synth] Batch: 1 request(s) [Registry] Qwen3-Embedding-0.6B-BF16.gguf -> Text-Enc [Registry] Qwen3-Embedding-0.6B-Q8_0.gguf -> Text-Enc [Registry] acestep-5Hz-lm-0.6B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-0.6B-Q8_0.gguf -> LM [Registry] acestep-5Hz-lm-1.7B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-1.7B-Q8_0.gguf -> LM [Registry] acestep-5Hz-lm-4B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q5_K_M.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q6_K.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q8_0.gguf -> LM [Registry] acestep-v15-base-BF16.gguf -> DiT [Registry] acestep-v15-base-Q4_K_M.gguf -> DiT [Registry] acestep-v15-base-Q5_K_M.gguf -> DiT [Registry] acestep-v15-base-Q6_K.gguf -> DiT [Registry] acestep-v15-base-Q8_0.gguf -> DiT [Registry] acestep-v15-sft-BF16.gguf -> DiT [Registry] acestep-v15-sft-Q4_K_M.gguf -> DiT [Registry] acestep-v15-sft-Q5_K_M.gguf -> DiT [Registry] acestep-v15-sft-Q6_K.gguf -> DiT [Registry] acestep-v15-sft-Q8_0.gguf -> DiT [Registry] acestep-v15-sftturbo50-BF16.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q4_K_M.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q5_K_M.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q6_K.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-BF16.gguf -> DiT [Registry] acestep-v15-turbo-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-continuous-BF16.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-shift1-BF16.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-shift3-BF16.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-base-BF16.gguf -> DiT [Registry] acestep-v15-xl-base-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-base-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-base-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-base-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-sft-BF16.gguf -> DiT [Registry] acestep-v15-xl-sft-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-sft-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-sft-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-sft-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-BF16.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-turbo-BF16.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q8_0.gguf -> DiT [Registry] scragvae-BF16.gguf -> VAE [Registry] vae-BF16.gguf -> VAE [Store] Created (policy=STRICT) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-turbo-Q5_K_M.gguf: 678 tensors, data at offset 56864 [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-turbo-Q5_K_M.gguf: 678 tensors, data at offset 56864 [Synth-Load] Ready: turbo=yes, fa=no, batch_cfg=yes [Resolve-Params] max audio codes across batch: 434 (86.8s @ 5Hz) [Resolve-T] T=2170, S=1085 [Resolve-T] seed=42, steps=8, guidance=1.0, shift=3.0, duration=88.0s [BPE] Loaded from GGUF: 151643 vocab, 151387 merges ggml_cuda_init: found 1 CUDA devices (Total VRAM: 97247 MiB): Device 0: NVIDIA RTX PRO 6000 Blackwell Workstation Edition, compute capability 12.0, VMM: yes, VRAM: 97247 MiB load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so ggml_vulkan: Found 1 Vulkan devices: ggml_vulkan: 0 = NVIDIA RTX PRO 6000 Blackwell Workstation Edition (NVIDIA) | uma: 0 | fp16: 1 | bf16: 0 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2 load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] TextEncoder backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/Qwen3-Embedding-0.6B-BF16.gguf: 310 tensors, data at offset 5337568 [Load] TextEncoder: 28L, H=1024, Nh=16/8 [Qwen3] Attn: Q+K+V fused [Qwen3] MLP: gate+up fused [WeightCtx] Loaded 310 tensors, 1136.5 MB into backend [Store] Load TextEnc: 340 ms [Debug] text_hidden: [70, 1024] first4: 3.705836 2.395382 0.221845 -13.145830 [Debug] lyric_embed: [167, 1024] first4: 0.029175 0.032227 -0.022339 -0.028809 [Store] Unload TextEnc (1136.5 MB) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] CondEncoder backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-turbo-Q5_K_M.gguf: 678 tensors, data at offset 56864 [Load] LyricEncoder: 8L [Qwen3] Attn: Q+K fused, V separate [Qwen3] MLP: gate+up fused [Load] TimbreEncoder: 4L [Qwen3] Attn: Q+K fused, V separate [Qwen3] MLP: gate+up fused [WeightCtx] Loaded 140 tensors, 412.5 MB into backend [Load] CondEncoder: lyric(8L), timbre(4L), text_proj, null_cond [Store] Load CondEnc: 717 ms [CondEnc] Lyric sliding mask: 167x167, window=128 [CondEnc] Timbre sliding mask: 1x1, window=128 [Encode] Packed: lyric=167 + timbre=1 + text=70 = 238 tokens [Encode-Text Batch0] 70+167 tokens -> enc_S=238, 16.8 ms [Debug] enc_hidden: [238, 2048] first4: 1.760519 -0.051711 -0.132143 0.058083 [Store] Unload CondEnc (412.5 MB) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] Detokenizer backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-turbo-Q5_K_M.gguf: 678 tensors, data at offset 56864 [WeightCtx] Loaded 30 tensors, 73.2 MB into backend [Load] Detokenizer: FSQ(6->2048) + 2L encoder(S=5, 2048->64) [Store] Load FSQ-Detok: 59 ms [Context] Decoded: 434 codes -> 2170 frames (86.8s @ 25Hz) [Build-Context Batch0] Detokenizer: 169.8 ms, 434 codes [Debug] detok_output: [2170, 64] first4: -0.126205 1.455492 0.291894 -0.650703 [Store] Unload FSQ-Detok (73.2 MB) [Init-Noise Batch0] Philox noise seed=42, [2170, 64] solver=euler [Debug] noise: [2170, 64] first4: 0.194336 2.156250 -0.171875 0.847656 [Debug] context: [2170, 128] first4: -0.126205 1.455492 0.291894 -0.650703 [Init-Noise] Starting: T=2170, S=1085, enc_S=238, steps=8, batch=1 (cover) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] DiT backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-turbo-Q5_K_M.gguf: 678 tensors, data at offset 56864 [DiT] Self-attn: Q+K fused, V separate [DiT] Cross-attn: all separate [DiT] MLP: gate+up fused [Load] null_condition_emb found (CFG available) [WeightCtx] Loaded 478 tensors, 1061.2 MB into backend [Load] DiT: 24 layers, H=2048, Nh=16/8, D=128 [Store] Load DiT: 819 ms [DiT] Batch N=1, T=2170, S=1085, enc_S=238 [DiT] Graph: 2063 nodes [DiT] Solver: ODE Euler (1 NFE/step, order 1) [Debug] tproj: [12288] first4: 0.260409 -0.161609 -0.102203 0.051602 [Debug] temb: [2048] first4: -0.000151 -0.132293 -0.035516 0.064751 [Debug] temb_t: [2048] first4: 0.000578 0.026708 -0.052786 0.063514 [Debug] temb_r: [2048] first4: -0.000729 -0.159001 0.017269 0.001237 [Debug] sinusoidal_t: [256] first4: 0.562486 0.789701 0.439822 -0.023583 [Debug] sinusoidal_r: [256] first4: 1.000000 1.000000 1.000000 1.000000 [Debug] temb_lin1_t: [2048] first4: -0.051153 -0.053631 -0.012192 -0.039024 [Debug] temb_lin1_r: [2048] first4: -0.016165 -0.021121 -0.015801 -0.000525 [Debug] hidden_after_proj_in: [2048, 1085] first4: -1.043457 -0.948303 0.536377 0.453583 [Debug] proj_in_input: [192, 2170] first4: -0.126205 1.455492 0.291894 -0.650703 [Debug] enc_after_cond_emb: [2048, 238] first4: -0.155945 0.748459 0.320129 -0.524803 [Debug] layer0_sa_input: [2048, 1085] first4: -0.721759 -0.751600 -0.052395 0.263968 [Debug] layer0_q_after_rope: [128, 16] first4: -0.155945 0.748459 0.320129 -0.524803 [Debug] layer0_k_after_rope: [128, 8] first4: -1.540865 -1.013701 0.146717 0.464547 [Debug] layer0_sa_output: [2048, 1085] first4: -1.498047 0.148193 -0.400635 0.486572 [Debug] layer0_attn_out: [2048, 1085] first4: -1.540865 -1.013701 0.146717 0.464547 [Debug] layer0_after_self_attn: [2048, 1085] first4: -1.540865 -1.013701 0.146717 0.464547 [Debug] layer0_after_cross_attn: [2048, 1085] first4: -1.578524 -0.769561 -0.336682 0.502175 [Debug] hidden_after_layer0: [2048, 1085] first4: -9.066914 0.597754 50.600498 -0.869583 [Debug] hidden_after_layer6: [2048, 1085] first4: -20.455296 0.417641 30.599295 -4.776403 [Debug] hidden_after_layer12: [2048, 1085] first4: -17.730957 -17.530033 73.683121 26.529156 [Debug] hidden_after_layer18: [2048, 1085] first4: -25.572363 10.031152 66.047440 18.983101 [Debug] hidden_after_layer23: [2048, 1085] first4: -4.296446 34.909538 200.473526 158.418533 [Debug] dit_step0_vt: [2170, 64] first4: 0.018551 1.147247 0.338320 2.330933 [Debug] dit_step0_xt: [2170, 64] first4: 0.193493 2.104102 -0.187253 0.741705 [DiT] Step 1/8 t=1.000 [Debug] dit_step1_vt: [2170, 64] first4: -0.213482 1.345932 -0.165329 1.875256 [Debug] dit_step1_xt: [2170, 64] first4: 0.205137 2.030688 -0.178235 0.639418 [DiT] Step 2/8 t=0.955 [Debug] dit_step2_vt: [2170, 64] first4: -0.048391 1.250273 0.054047 2.355942 [Debug] dit_step2_xt: [2170, 64] first4: 0.208363 1.947336 -0.181838 0.482355 [DiT] Step 3/8 t=0.900 [Debug] dit_step3_vt: [2170, 64] first4: 0.272675 1.097305 0.221283 2.595299 [Debug] dit_step3_xt: [2170, 64] first4: 0.185640 1.855894 -0.200279 0.266080 [DiT] Step 4/8 t=0.833 [Debug] dit_step4_vt: [2170, 64] first4: 0.311768 0.923492 0.056274 2.677630 [Debug] dit_step4_xt: [2170, 64] first4: 0.152237 1.756949 -0.206308 -0.020809 [DiT] Step 5/8 t=0.750 [Debug] dit_step5_vt: [2170, 64] first4: 0.334290 0.794403 -0.324097 2.681274 [Debug] dit_step5_xt: [2170, 64] first4: 0.104481 1.643462 -0.160008 -0.403848 [DiT] Step 6/8 t=0.643 [Debug] dit_step6_vt: [2170, 64] first4: 0.204941 0.596207 -0.831627 2.757751 [Debug] dit_step6_xt: [2170, 64] first4: 0.063493 1.524221 0.006317 -0.955398 [DiT] Step 7/8 t=0.500 [Debug] dit_step7_vt: [2170, 64] first4: -0.047699 0.263275 -1.311249 3.134644 [Debug] dit_x0: [2170, 64] first4: 0.077803 1.445238 0.399692 -1.895791 [DiT] Step 8/8 t=0.300 [DiT-Generate] Total: 290.5 ms (290.5 ms/sample) [Debug] dit_output: [2170, 64] first4: 0.077803 1.445238 0.399692 -1.895791 [Store] Unload DiT (1061.2 MB) [GGUF] /mnt/workspace/git/acestep.cpp/models/vae-BF16.gguf: 365 tensors, data at offset 30048 load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] VAE backend: Vulkan0 (CPU threads: 16) [VAE] Backend: Vulkan0, Weight buffer: 161.1 MB [VAE] Loaded: 5 blocks, upsample=1920x, F32 activations [Store] Load VAE-Dec: 669 ms [VAE] Tiled decode: 3 tiles (chunk=1024, overlap=64, stride=896) [VAE] Graph: 479 nodes, T_latent=960 [VAE] Upsample factor: 1920.00 (expected ~1920) [VAE] Graph: 479 nodes, T_latent=1024 [VAE] Graph: 479 nodes, T_latent=442 [VAE] Tiled decode done: 3 tiles -> T_audio=4166400 (86.80s @ 48kHz) [VAE-Decode Batch0] Decode: 947.6 ms [Debug] vae_audio: [2, 4166400] first4: 0.000963 0.001694 0.001477 0.002009 [Store] Unload VAE-Dec (161.1 MB) [MP3] Encoding 86.8s @ 128 kbps, 48000 Hz stereo [MP3] 1388928 bytes (12.0:1), 449 ms (193.31x realtime), 32 threads [MP3] Wrote ggml-turbo/request00.mp3 [Ace-Synth] All done 2026-07-12 10:22:18.378 | DEBUG | acestep.core.generation.handler.init_service_memory_basic:_apply_malloc_mmap_threshold:50 - [memory] Set M_MMAP_THRESHOLD=131072 for immediate OS reclaim of large frees Skipping import of cpp extensions due to incompatible torch version. Please upgrade to torch >= 2.11.0 (found 2.10.0+cu128). 2026-07-12 10:22:20.148 | WARNING | acestep.training.trainer::40 - bitsandbytes not installed. Using standard AdamW. 2026-07-12 10:22:20.150 | INFO | acestep.gpu_config:get_gpu_memory_gb:578 - CUDA GPU detected: NVIDIA RTX PRO 6000 Blackwell Workstation Edition (95.0 GB) 2026-07-12 10:22:20.219 | INFO | acestep.core.generation.handler.init_service_loader:_load_main_model_from_checkpoint:174 - [initialize_service] Attempting to load model with attention implementation: sdpa Unable to import `torchao` Tensor objects. This may affect loading checkpoints serialized with `torchao` 2026-07-12 10:22:21.836 | INFO | acestep.core.generation.handler.generate_music:generate_music:289 - [generate_music] Turbo model detected: overriding guidance_scale 0.0 -> 1.0 (turbo does not use CFG). 2026-07-12 10:22:21.837 | INFO | acestep.core.generation.handler.generate_music:generate_music:304 - [generate_music] Starting generation... 2026-07-12 10:22:21.837 | INFO | acestep.core.generation.handler.generate_music:generate_music:307 - [generate_music] Preparing inputs... 2026-07-12 10:22:21.837 | INFO | acestep.core.generation.handler.generate_music_request:_prepare_reference_and_source_audio:179 - [generate_music] cover task: no src_audio but audio codes provided, proceeding with codes 2026-07-12 10:22:21.840 | INFO | acestep.core.generation.handler.generate_music:_vram_preflight_check:156 - [generate_music] VRAM pre-flight: 87.83 GB free, ~0.94 GB needed (batch=1, duration=88s, mode=turbo). 2026-07-12 10:22:21.845 | INFO | acestep.core.generation.handler.conditioning_target:_prepare_target_latents_and_wavs:78 - [generate_music] Decoding audio codes for item 0... 2026-07-12 10:22:22.039 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_precomputed_lm_hints:31 - [generate_music] Decoding audio codes for LM hints for item 0... 2026-07-12 10:22:22.041 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:119 - ====================================================================== 2026-07-12 10:22:22.041 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:120 - πŸ” [DEBUG] DiT TEXT ENCODER INPUT (Inference) 2026-07-12 10:22:22.041 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:121 - ====================================================================== 2026-07-12 10:22:22.041 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:122 - text_prompt: # Instruction Generate audio semantic tokens based on the given conditions: # Caption An upbeat and anthemic pop-rock track driven by bright, slightly overdriven # Metas - bpm: 83 - timesignature: 4 - keyscale: G major - duration: 88 seconds <|endoftext|> 2026-07-12 10:22:22.041 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:123 - ====================================================================== 2026-07-12 10:22:22.041 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:124 - lyrics_text: # Languages fr # Lyric # Lyric [Intro - Guitar Riff] [Verse 1] Dans le monde des tutos virtuels G ta toise en nouvelle passion Avec Ggendoline et PumbΓ© Γ  midi La communautΓ©, c'est l'unitΓ© Quel joie, une clΓ© [Chorus] Dans le monde des tutos virtuels GΓ’ndoline et PumbΓ© Γ  midi Une famille Γ  connecter, c'est vrai D'un enfant qui voit toi fusionner [Guitar Solo] [Verse 2] Dans le monde des tutos virtuels GΓ’ndoline, PumbΓ© Γ  midi Une famille Γ  connecter, c'est vrai D'un enfant qui voit toi fusionner<|endoftext|> 2026-07-12 10:22:22.041 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:125 - ====================================================================== 2026-07-12 10:22:22.049 | INFO | acestep.core.generation.handler.conditioning_embed:preprocess_batch:110 - [preprocess_batch] Inferring prompt embeddings... 2026-07-12 10:22:22.063 | INFO | acestep.core.generation.handler.conditioning_embed:preprocess_batch:113 - [preprocess_batch] Inferring lyric embeddings... 2026-07-12 10:22:22.063 | INFO | acestep.core.generation.handler.service_generate_execute:_execute_service_generate_diffusion:161 - [service_generate] Generating audio... (DiT backend: PyTorch (cuda)) 2026-07-12 10:22:22.089 | INFO | acestep.core.generation.handler.service_generate_execute:_execute_service_generate_diffusion:263 - [service_generate] DiT diffusion via PyTorch (cuda)... 2026-07-12 10:22:22.097 | INFO | acestep.models.common.dcw_correction:__init__:98 - [DCW] Active β€” mode=double, scaler=0.0500, high_scaler=0.0200, wavelet='haar' 2026-07-12 10:22:22.186 | WARNING | acestep.models.common.dcw_loader:_try_import:39 - DCW is enabled but 'pytorch_wavelets' is not installed. Install with `pip install pytorch_wavelets PyWavelets` to use Differential Correction in Wavelet domain. Falling back to no-op for this generation. 2026-07-12 10:22:22.414 | INFO | acestep.core.generation.handler.generate_music_decode:_prepare_generate_music_decode_state:42 - [generate_music] Model generation completed. Decoding latents... 2026-07-12 10:22:22.415 | DEBUG | acestep.core.generation.handler.generate_music_decode:_prepare_generate_music_decode_state:64 - [generate_music] pred_latents: torch.Size([1, 2170, 64]), dtype=torch.bfloat16 2026-07-12 10:22:22.415 | DEBUG | acestep.core.generation.handler.generate_music_decode:_prepare_generate_music_decode_state:65 - [generate_music] time_costs: {'encoder_time_cost': 0.006894826889038086, 'diffusion_time_cost': 0.31703972816467285, 'diffusion_per_step_time_cost': 0.039629966020584106, 'total_time_cost': 0.32393455505371094, 'offload_time_cost': 0.0} 2026-07-12 10:22:22.429 | INFO | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:127 - [generate_music] Decoding latents with VAE... 2026-07-12 10:22:22.431 | DEBUG | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:136 - [generate_music] Before VAE decode: allocated=5.98GB, max=7.29GB 2026-07-12 10:22:22.431 | INFO | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:154 - [generate_music] Effective free VRAM before VAE decode: 87.73 GB 2026-07-12 10:22:22.431 | INFO | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:172 - [generate_music] Using tiled VAE decode to reduce VRAM usage... 2026-07-12 10:22:22.431 | DEBUG | acestep.core.generation.handler.memory_utils:_get_auto_decode_chunk_size:108 - [_get_auto_decode_chunk_size] Effective free VRAM: 87.73 GB 2026-07-12 10:22:22.432 | DEBUG | acestep.core.generation.handler.memory_utils:_should_offload_wav_to_cpu:131 - [_should_offload_wav_to_cpu] Effective free VRAM: 87.73 GB 2026-07-12 10:22:22.432 | INFO | acestep.core.generation.handler.vae_decode:tiled_decode:56 - [tiled_decode] chunk_size=512, offload_wav_to_cpu=False, latents_shape=torch.Size([1, 64, 2170]) 2026-07-12 10:22:22.708 | DEBUG | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:193 - [generate_music] After VAE decode: allocated=6.15GB, max=7.44GB 2026-07-12 10:22:22.868 | INFO | acestep.core.generation.handler.generate_music_payload:_build_generate_music_success_payload:38 - [generate_music] VAE decode completed. Preparing audio tensors... 2026-07-12 10:22:22.871 | INFO | acestep.core.generation.handler.generate_music_payload:_build_generate_music_success_payload:50 - [generate_music] Done! Generated 1 audio tensors. [Request] Parsed ggml-sft/request0.json [Request] seed=42 lm_batch=1 synth_batch=1 [Request] caption: An upbeat and anthemic pop-rock track driven by bright, slig... [Request] lyrics: 490 bytes [Request] bpm=83 dur=88 key=G major ts=4 lang=fr [Request] lm: temp=0.85 cfg=2.0 top_p=0.90 top_k=0 [Request] dit: steps=50 guidance=1.0 shift=1.0 [Request] task_type: text2music [Request] solver: euler (stork_substeps=10) [Request] lm_mode: generate [Request] output_format: mp3 [Request] synth_model: acestep-v15-sft-Q5_K_M.gguf [Request] vae: vae-BF16.gguf [Request] audio_codes: (present) [Ace-Synth] Batch: 1 request(s) [Registry] Qwen3-Embedding-0.6B-BF16.gguf -> Text-Enc [Registry] Qwen3-Embedding-0.6B-Q8_0.gguf -> Text-Enc [Registry] acestep-5Hz-lm-0.6B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-0.6B-Q8_0.gguf -> LM [Registry] acestep-5Hz-lm-1.7B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-1.7B-Q8_0.gguf -> LM [Registry] acestep-5Hz-lm-4B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q5_K_M.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q6_K.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q8_0.gguf -> LM [Registry] acestep-v15-base-BF16.gguf -> DiT [Registry] acestep-v15-base-Q4_K_M.gguf -> DiT [Registry] acestep-v15-base-Q5_K_M.gguf -> DiT [Registry] acestep-v15-base-Q6_K.gguf -> DiT [Registry] acestep-v15-base-Q8_0.gguf -> DiT [Registry] acestep-v15-sft-BF16.gguf -> DiT [Registry] acestep-v15-sft-Q4_K_M.gguf -> DiT [Registry] acestep-v15-sft-Q5_K_M.gguf -> DiT [Registry] acestep-v15-sft-Q6_K.gguf -> DiT [Registry] acestep-v15-sft-Q8_0.gguf -> DiT [Registry] acestep-v15-sftturbo50-BF16.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q4_K_M.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q5_K_M.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q6_K.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-BF16.gguf -> DiT [Registry] acestep-v15-turbo-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-continuous-BF16.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-shift1-BF16.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-shift3-BF16.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-base-BF16.gguf -> DiT [Registry] acestep-v15-xl-base-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-base-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-base-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-base-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-sft-BF16.gguf -> DiT [Registry] acestep-v15-xl-sft-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-sft-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-sft-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-sft-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-BF16.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-turbo-BF16.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q8_0.gguf -> DiT [Registry] scragvae-BF16.gguf -> VAE [Registry] vae-BF16.gguf -> VAE [Store] Created (policy=STRICT) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-sft-Q5_K_M.gguf: 678 tensors, data at offset 56800 [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-sft-Q5_K_M.gguf: 678 tensors, data at offset 56800 [Synth-Load] Ready: turbo=no, fa=no, batch_cfg=yes [Resolve-Params] max audio codes across batch: 434 (86.8s @ 5Hz) [Resolve-T] T=2170, S=1085 [Resolve-T] seed=42, steps=50, guidance=1.0, shift=1.0, duration=88.0s [BPE] Loaded from GGUF: 151643 vocab, 151387 merges ggml_cuda_init: found 1 CUDA devices (Total VRAM: 97247 MiB): Device 0: NVIDIA RTX PRO 6000 Blackwell Workstation Edition, compute capability 12.0, VMM: yes, VRAM: 97247 MiB load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so ggml_vulkan: Found 1 Vulkan devices: ggml_vulkan: 0 = NVIDIA RTX PRO 6000 Blackwell Workstation Edition (NVIDIA) | uma: 0 | fp16: 1 | bf16: 0 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2 load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] TextEncoder backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/Qwen3-Embedding-0.6B-BF16.gguf: 310 tensors, data at offset 5337568 [Load] TextEncoder: 28L, H=1024, Nh=16/8 [Qwen3] Attn: Q+K+V fused [Qwen3] MLP: gate+up fused [WeightCtx] Loaded 310 tensors, 1136.5 MB into backend [Store] Load TextEnc: 354 ms [Debug] text_hidden: [70, 1024] first4: 3.705836 2.395382 0.221845 -13.145830 [Debug] lyric_embed: [167, 1024] first4: 0.029175 0.032227 -0.022339 -0.028809 [Store] Unload TextEnc (1136.5 MB) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] CondEncoder backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-sft-Q5_K_M.gguf: 678 tensors, data at offset 56800 [Load] LyricEncoder: 8L [Qwen3] Attn: Q+K fused, V separate [Qwen3] MLP: gate+up fused [Load] TimbreEncoder: 4L [Qwen3] Attn: Q+K fused, V separate [Qwen3] MLP: gate+up fused [WeightCtx] Loaded 140 tensors, 412.5 MB into backend [Load] CondEncoder: lyric(8L), timbre(4L), text_proj, null_cond [Store] Load CondEnc: 700 ms [CondEnc] Lyric sliding mask: 167x167, window=128 [CondEnc] Timbre sliding mask: 1x1, window=128 [Encode] Packed: lyric=167 + timbre=1 + text=70 = 238 tokens [Encode-Text Batch0] 70+167 tokens -> enc_S=238, 8.2 ms [Debug] enc_hidden: [238, 2048] first4: 1.760519 -0.051711 -0.132143 0.058083 [Store] Unload CondEnc (412.5 MB) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] Detokenizer backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-sft-Q5_K_M.gguf: 678 tensors, data at offset 56800 [WeightCtx] Loaded 30 tensors, 73.2 MB into backend [Load] Detokenizer: FSQ(6->2048) + 2L encoder(S=5, 2048->64) [Store] Load FSQ-Detok: 53 ms [Context] Decoded: 434 codes -> 2170 frames (86.8s @ 25Hz) [Build-Context Batch0] Detokenizer: 188.8 ms, 434 codes [Debug] detok_output: [2170, 64] first4: -0.126205 1.455492 0.291894 -0.650703 [Store] Unload FSQ-Detok (73.2 MB) [Init-Noise Batch0] Philox noise seed=42, [2170, 64] solver=euler [Debug] noise: [2170, 64] first4: 0.194336 2.156250 -0.171875 0.847656 [Debug] context: [2170, 128] first4: -0.126205 1.455492 0.291894 -0.650703 [Init-Noise] Starting: T=2170, S=1085, enc_S=238, steps=50, batch=1 (cover) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] DiT backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-sft-Q5_K_M.gguf: 678 tensors, data at offset 56800 [DiT] Self-attn: Q+K fused, V separate [DiT] Cross-attn: all separate [DiT] MLP: gate+up fused [Load] null_condition_emb found (CFG available) [WeightCtx] Loaded 478 tensors, 1061.2 MB into backend [Load] DiT: 24 layers, H=2048, Nh=16/8, D=128 [Store] Load DiT: 810 ms [DiT] Batch N=1, T=2170, S=1085, enc_S=238 [DiT] Graph: 2063 nodes [DiT] Solver: ODE Euler (1 NFE/step, order 1) [Debug] tproj: [12288] first4: 0.155137 -0.118305 -0.084248 0.082623 [Debug] temb: [2048] first4: -0.002843 -0.176820 0.004745 -0.001924 [Debug] temb_t: [2048] first4: -0.001351 0.003023 -0.012552 -0.001712 [Debug] temb_r: [2048] first4: -0.001491 -0.179843 0.017298 -0.000212 [Debug] sinusoidal_t: [256] first4: 0.562486 0.789701 0.439822 -0.023583 [Debug] sinusoidal_r: [256] first4: 1.000000 1.000000 1.000000 1.000000 [Debug] temb_lin1_t: [2048] first4: -0.034754 0.028817 0.027120 -0.031729 [Debug] temb_lin1_r: [2048] first4: -0.002680 0.004202 0.000655 -0.002088 [Debug] hidden_after_proj_in: [2048, 1085] first4: -1.093750 -0.904541 0.498291 0.448029 [Debug] proj_in_input: [192, 2170] first4: -0.126205 1.455492 0.291894 -0.650703 [Debug] enc_after_cond_emb: [2048, 238] first4: -0.175995 0.837418 0.352905 -0.564758 [Debug] layer0_sa_input: [2048, 1085] first4: -0.918976 -0.705216 -0.034956 0.298489 [Debug] layer0_q_after_rope: [128, 16] first4: -0.175995 0.837418 0.352905 -0.564758 [Debug] layer0_k_after_rope: [128, 8] first4: -1.804577 -1.343851 -0.172177 0.389113 [Debug] layer0_sa_output: [2048, 1085] first4: -1.642578 0.773438 -0.594727 0.532715 [Debug] layer0_attn_out: [2048, 1085] first4: -1.804577 -1.343851 -0.172177 0.389113 [Debug] layer0_after_self_attn: [2048, 1085] first4: -1.804577 -1.343851 -0.172177 0.389113 [Debug] layer0_after_cross_attn: [2048, 1085] first4: -1.961926 -0.976175 -0.429502 0.408568 [Debug] hidden_after_layer0: [2048, 1085] first4: -9.510232 1.328141 57.980049 -1.647067 [Debug] hidden_after_layer6: [2048, 1085] first4: -14.639507 3.598928 57.444698 -0.184403 [Debug] hidden_after_layer12: [2048, 1085] first4: -12.856318 -0.021215 -23.480938 -0.564669 [Debug] hidden_after_layer18: [2048, 1085] first4: -5.399233 15.654142 -42.618336 -9.613000 [Debug] hidden_after_layer23: [2048, 1085] first4: 32.141037 61.246735 49.951061 -14.661944 [Debug] dit_step0_vt: [2170, 64] first4: -0.641663 2.515259 -0.197014 1.444469 [Debug] dit_step0_xt: [2170, 64] first4: 0.207169 2.105945 -0.167935 0.818767 [DiT] Step 1/50 t=1.000 [Debug] dit_step1_vt: [2170, 64] first4: -0.704742 2.528320 -0.057358 1.249832 [Debug] dit_step1_xt: [2170, 64] first4: 0.221264 2.055378 -0.166788 0.793770 [DiT] Step 2/50 t=0.980 [Debug] dit_step2_vt: [2170, 64] first4: -0.701294 2.486572 -0.009857 1.276260 [Debug] dit_step2_xt: [2170, 64] first4: 0.235290 2.005647 -0.166590 0.768245 [DiT] Step 3/50 t=0.960 [Debug] dit_step3_vt: [2170, 64] first4: -0.683594 2.418701 0.031006 1.296326 [Debug] dit_step3_xt: [2170, 64] first4: 0.248962 1.957273 -0.167211 0.742319 [DiT] Step 4/50 t=0.940 [Debug] dit_step4_vt: [2170, 64] first4: -0.617249 2.334717 0.029877 1.296448 [Debug] dit_step4_xt: [2170, 64] first4: 0.261307 1.910579 -0.167808 0.716389 [DiT] Step 5/50 t=0.920 [Debug] dit_step5_vt: [2170, 64] first4: -0.572449 2.253174 0.010483 1.298309 [Debug] dit_step5_xt: [2170, 64] first4: 0.272756 1.865515 -0.168018 0.690423 [DiT] Step 6/50 t=0.900 [Debug] dit_step6_vt: [2170, 64] first4: -0.537292 2.204224 -0.006714 1.289093 [Debug] dit_step6_xt: [2170, 64] first4: 0.283502 1.821431 -0.167883 0.664641 [DiT] Step 7/50 t=0.880 [Debug] dit_step7_vt: [2170, 64] first4: -0.508911 2.141724 -0.024338 1.287399 [Debug] dit_step7_xt: [2170, 64] first4: 0.293680 1.778596 -0.167397 0.638894 [DiT] Step 8/50 t=0.860 [Debug] dit_step8_vt: [2170, 64] first4: -0.462341 2.061462 -0.033276 1.282303 [Debug] dit_step8_xt: [2170, 64] first4: 0.302927 1.737367 -0.166731 0.613247 [DiT] Step 9/50 t=0.840 [Debug] dit_step9_vt: [2170, 64] first4: -0.402283 1.975891 -0.021896 1.272064 [Debug] dit_step9_xt: [2170, 64] first4: 0.310972 1.697849 -0.166293 0.587806 [DiT] Step 10/50 t=0.820 [Debug] dit_step10_vt: [2170, 64] first4: -0.350891 1.893738 -0.006638 1.253418 [Debug] dit_step10_xt: [2170, 64] first4: 0.317990 1.659974 -0.166161 0.562738 [DiT] Step 11/50 t=0.800 [Debug] dit_step11_vt: [2170, 64] first4: -0.317139 1.818420 0.000679 1.240738 [Debug] dit_step11_xt: [2170, 64] first4: 0.324333 1.623606 -0.166174 0.537923 [DiT] Step 12/50 t=0.780 [Debug] dit_step12_vt: [2170, 64] first4: -0.298706 1.752808 0.001938 1.228821 [Debug] dit_step12_xt: [2170, 64] first4: 0.330307 1.588550 -0.166213 0.513347 [DiT] Step 13/50 t=0.760 [Debug] dit_step13_vt: [2170, 64] first4: -0.286560 1.686890 -0.009232 1.211166 [Debug] dit_step13_xt: [2170, 64] first4: 0.336038 1.554812 -0.166028 0.489123 [DiT] Step 14/50 t=0.740 [Debug] dit_step14_vt: [2170, 64] first4: -0.266083 1.606354 -0.019852 1.196686 [Debug] dit_step14_xt: [2170, 64] first4: 0.341360 1.522685 -0.165631 0.465190 [DiT] Step 15/50 t=0.720 [Debug] dit_step15_vt: [2170, 64] first4: -0.266998 1.538376 -0.036957 1.192703 [Debug] dit_step15_xt: [2170, 64] first4: 0.346700 1.491918 -0.164892 0.441336 [DiT] Step 16/50 t=0.700 [Debug] dit_step16_vt: [2170, 64] first4: -0.264816 1.458122 -0.050293 1.183399 [Debug] dit_step16_xt: [2170, 64] first4: 0.351996 1.462755 -0.163886 0.417668 [DiT] Step 17/50 t=0.680 [Debug] dit_step17_vt: [2170, 64] first4: -0.263489 1.380098 -0.070801 1.171424 [Debug] dit_step17_xt: [2170, 64] first4: 0.357266 1.435153 -0.162470 0.394239 [DiT] Step 18/50 t=0.660 [Debug] dit_step18_vt: [2170, 64] first4: -0.282043 1.301178 -0.094391 1.163564 [Debug] dit_step18_xt: [2170, 64] first4: 0.362907 1.409130 -0.160582 0.370968 [DiT] Step 19/50 t=0.640 [Debug] dit_step19_vt: [2170, 64] first4: -0.291248 1.222473 -0.105896 1.152061 [Debug] dit_step19_xt: [2170, 64] first4: 0.368732 1.384680 -0.158464 0.347927 [DiT] Step 20/50 t=0.620 [Debug] dit_step20_vt: [2170, 64] first4: -0.304316 1.139389 -0.113770 1.128166 [Debug] dit_step20_xt: [2170, 64] first4: 0.374818 1.361892 -0.156189 0.325363 [DiT] Step 21/50 t=0.600 [Debug] dit_step21_vt: [2170, 64] first4: -0.326122 1.056488 -0.125977 1.116730 [Debug] dit_step21_xt: [2170, 64] first4: 0.381341 1.340762 -0.153670 0.303029 [DiT] Step 22/50 t=0.580 [Debug] dit_step22_vt: [2170, 64] first4: -0.346006 0.955841 -0.139191 1.087845 [Debug] dit_step22_xt: [2170, 64] first4: 0.388261 1.321646 -0.150886 0.281272 [DiT] Step 23/50 t=0.560 [Debug] dit_step23_vt: [2170, 64] first4: -0.369873 0.863525 -0.134552 1.069717 [Debug] dit_step23_xt: [2170, 64] first4: 0.395658 1.304375 -0.148195 0.259877 [DiT] Step 24/50 t=0.540 [Debug] dit_step24_vt: [2170, 64] first4: -0.399165 0.771606 -0.147583 1.045990 [Debug] dit_step24_xt: [2170, 64] first4: 0.403641 1.288943 -0.145243 0.238958 [DiT] Step 25/50 t=0.520 [Debug] dit_step25_vt: [2170, 64] first4: -0.427725 0.674744 -0.134216 1.022552 [Debug] dit_step25_xt: [2170, 64] first4: 0.412196 1.275448 -0.142559 0.218507 [DiT] Step 26/50 t=0.500 [Debug] dit_step26_vt: [2170, 64] first4: -0.456692 0.569366 -0.147980 0.992889 [Debug] dit_step26_xt: [2170, 64] first4: 0.421330 1.264061 -0.139599 0.198649 [DiT] Step 27/50 t=0.480 [Debug] dit_step27_vt: [2170, 64] first4: -0.479267 0.457214 -0.139862 0.976501 [Debug] dit_step27_xt: [2170, 64] first4: 0.430915 1.254917 -0.136802 0.179119 [DiT] Step 28/50 t=0.460 [Debug] dit_step28_vt: [2170, 64] first4: -0.507675 0.362152 -0.133667 0.943054 [Debug] dit_step28_xt: [2170, 64] first4: 0.441069 1.247674 -0.134128 0.160258 [DiT] Step 29/50 t=0.440 [Debug] dit_step29_vt: [2170, 64] first4: -0.538483 0.235397 -0.132172 0.918579 [Debug] dit_step29_xt: [2170, 64] first4: 0.451838 1.242966 -0.131485 0.141886 [DiT] Step 30/50 t=0.420 [Debug] dit_step30_vt: [2170, 64] first4: -0.543774 0.128906 -0.114914 0.894653 [Debug] dit_step30_xt: [2170, 64] first4: 0.462714 1.240387 -0.129187 0.123993 [DiT] Step 31/50 t=0.400 [Debug] dit_step31_vt: [2170, 64] first4: -0.552910 0.017872 -0.103806 0.884338 [Debug] dit_step31_xt: [2170, 64] first4: 0.473772 1.240030 -0.127111 0.106306 [DiT] Step 32/50 t=0.380 [Debug] dit_step32_vt: [2170, 64] first4: -0.571163 -0.080284 -0.086761 0.862000 [Debug] dit_step32_xt: [2170, 64] first4: 0.485195 1.241636 -0.125375 0.089066 [DiT] Step 33/50 t=0.360 [Debug] dit_step33_vt: [2170, 64] first4: -0.573780 -0.186707 -0.076267 0.832031 [Debug] dit_step33_xt: [2170, 64] first4: 0.496671 1.245370 -0.123850 0.072426 [DiT] Step 34/50 t=0.340 [Debug] dit_step34_vt: [2170, 64] first4: -0.590160 -0.291321 -0.057830 0.809265 [Debug] dit_step34_xt: [2170, 64] first4: 0.508474 1.251196 -0.122693 0.056240 [DiT] Step 35/50 t=0.320 [Debug] dit_step35_vt: [2170, 64] first4: -0.591011 -0.397430 -0.043015 0.777222 [Debug] dit_step35_xt: [2170, 64] first4: 0.520294 1.259145 -0.121833 0.040696 [DiT] Step 36/50 t=0.300 [Debug] dit_step36_vt: [2170, 64] first4: -0.592297 -0.493744 -0.020630 0.741821 [Debug] dit_step36_xt: [2170, 64] first4: 0.532140 1.269020 -0.121421 0.025860 [DiT] Step 37/50 t=0.280 [Debug] dit_step37_vt: [2170, 64] first4: -0.590017 -0.595718 0.016357 0.711609 [Debug] dit_step37_xt: [2170, 64] first4: 0.543940 1.280934 -0.121748 0.011627 [DiT] Step 38/50 t=0.260 [Debug] dit_step38_vt: [2170, 64] first4: -0.581253 -0.676712 0.037994 0.676819 [Debug] dit_step38_xt: [2170, 64] first4: 0.555566 1.294468 -0.122508 -0.001909 [DiT] Step 39/50 t=0.240 [Debug] dit_step39_vt: [2170, 64] first4: -0.577222 -0.769638 0.084686 0.643982 [Debug] dit_step39_xt: [2170, 64] first4: 0.567110 1.309861 -0.124201 -0.014789 [DiT] Step 40/50 t=0.220 [Debug] dit_step40_vt: [2170, 64] first4: -0.572479 -0.841080 0.110275 0.596954 [Debug] dit_step40_xt: [2170, 64] first4: 0.578560 1.326683 -0.126407 -0.026728 [DiT] Step 41/50 t=0.200 [Debug] dit_step41_vt: [2170, 64] first4: -0.559666 -0.925786 0.155632 0.560951 [Debug] dit_step41_xt: [2170, 64] first4: 0.589753 1.345198 -0.129520 -0.037947 [DiT] Step 42/50 t=0.180 [Debug] dit_step42_vt: [2170, 64] first4: -0.545708 -0.999874 0.200920 0.501526 [Debug] dit_step42_xt: [2170, 64] first4: 0.600667 1.365196 -0.133538 -0.047977 [DiT] Step 43/50 t=0.160 [Debug] dit_step43_vt: [2170, 64] first4: -0.532402 -1.064869 0.243469 0.449829 [Debug] dit_step43_xt: [2170, 64] first4: 0.611315 1.386493 -0.138407 -0.056974 [DiT] Step 44/50 t=0.140 [Debug] dit_step44_vt: [2170, 64] first4: -0.513874 -1.124847 0.278427 0.381577 [Debug] dit_step44_xt: [2170, 64] first4: 0.621592 1.408990 -0.143976 -0.064605 [DiT] Step 45/50 t=0.120 [Debug] dit_step45_vt: [2170, 64] first4: -0.498703 -1.172821 0.300079 0.327286 [Debug] dit_step45_xt: [2170, 64] first4: 0.631567 1.432446 -0.149977 -0.071151 [DiT] Step 46/50 t=0.100 [Debug] dit_step46_vt: [2170, 64] first4: -0.468637 -1.227295 0.325722 0.276005 [Debug] dit_step46_xt: [2170, 64] first4: 0.640939 1.456992 -0.156492 -0.076671 [DiT] Step 47/50 t=0.080 [Debug] dit_step47_vt: [2170, 64] first4: -0.442764 -1.257843 0.321384 0.233047 [Debug] dit_step47_xt: [2170, 64] first4: 0.649795 1.482149 -0.162920 -0.081332 [DiT] Step 48/50 t=0.060 [Debug] dit_step48_vt: [2170, 64] first4: -0.408478 -1.292183 0.311249 0.195480 [Debug] dit_step48_xt: [2170, 64] first4: 0.657964 1.507993 -0.169145 -0.085242 [DiT] Step 49/50 t=0.040 [Debug] dit_step49_vt: [2170, 64] first4: -0.448029 -1.331593 0.293915 0.189575 [Debug] dit_x0: [2170, 64] first4: 0.666925 1.534625 -0.175023 -0.089033 [DiT] Step 50/50 t=0.020 [DiT-Generate] Total: 1634.0 ms (1634.0 ms/sample) [Debug] dit_output: [2170, 64] first4: 0.666925 1.534625 -0.175023 -0.089033 [Store] Unload DiT (1061.2 MB) [GGUF] /mnt/workspace/git/acestep.cpp/models/vae-BF16.gguf: 365 tensors, data at offset 30048 load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] VAE backend: Vulkan0 (CPU threads: 16) [VAE] Backend: Vulkan0, Weight buffer: 161.1 MB [VAE] Loaded: 5 blocks, upsample=1920x, F32 activations [Store] Load VAE-Dec: 665 ms [VAE] Tiled decode: 3 tiles (chunk=1024, overlap=64, stride=896) [VAE] Graph: 479 nodes, T_latent=960 [VAE] Upsample factor: 1920.00 (expected ~1920) [VAE] Graph: 479 nodes, T_latent=1024 [VAE] Graph: 479 nodes, T_latent=442 [VAE] Tiled decode done: 3 tiles -> T_audio=4166400 (86.80s @ 48kHz) [VAE-Decode Batch0] Decode: 946.4 ms [Debug] vae_audio: [2, 4166400] first4: -0.001466 -0.000953 -0.001243 -0.000814 [Store] Unload VAE-Dec (161.1 MB) [MP3] Encoding 86.8s @ 128 kbps, 48000 Hz stereo [MP3] 1388928 bytes (12.0:1), 404 ms (214.76x realtime), 32 threads [MP3] Wrote ggml-sft/request00.mp3 [Ace-Synth] All done 2026-07-12 10:22:29.767 | INFO | acestep.core.generation.handler.init_service_loader:_load_main_model_from_checkpoint:174 - [initialize_service] Attempting to load model with attention implementation: sdpa 2026-07-12 10:22:30.886 | INFO | acestep.core.generation.handler.generate_music:generate_music:304 - [generate_music] Starting generation... 2026-07-12 10:22:30.886 | INFO | acestep.core.generation.handler.generate_music:generate_music:307 - [generate_music] Preparing inputs... 2026-07-12 10:22:30.886 | INFO | acestep.core.generation.handler.generate_music_request:_prepare_reference_and_source_audio:179 - [generate_music] cover task: no src_audio but audio codes provided, proceeding with codes 2026-07-12 10:22:30.889 | INFO | acestep.core.generation.handler.generate_music:_vram_preflight_check:156 - [generate_music] VRAM pre-flight: 81.69 GB free, ~0.94 GB needed (batch=1, duration=88s, mode=turbo). 2026-07-12 10:22:30.894 | INFO | acestep.core.generation.handler.conditioning_target:_prepare_target_latents_and_wavs:78 - [generate_music] Decoding audio codes for item 0... 2026-07-12 10:22:30.898 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_precomputed_lm_hints:31 - [generate_music] Decoding audio codes for LM hints for item 0... 2026-07-12 10:22:30.900 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:119 - ====================================================================== 2026-07-12 10:22:30.900 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:120 - πŸ” [DEBUG] DiT TEXT ENCODER INPUT (Inference) 2026-07-12 10:22:30.900 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:121 - ====================================================================== 2026-07-12 10:22:30.900 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:122 - text_prompt: # Instruction Generate audio semantic tokens based on the given conditions: # Caption An upbeat and anthemic pop-rock track driven by bright, slightly overdriven # Metas - bpm: 83 - timesignature: 4 - keyscale: G major - duration: 88 seconds <|endoftext|> 2026-07-12 10:22:30.900 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:123 - ====================================================================== 2026-07-12 10:22:30.900 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:124 - lyrics_text: # Languages fr # Lyric # Lyric [Intro - Guitar Riff] [Verse 1] Dans le monde des tutos virtuels G ta toise en nouvelle passion Avec Ggendoline et PumbΓ© Γ  midi La communautΓ©, c'est l'unitΓ© Quel joie, une clΓ© [Chorus] Dans le monde des tutos virtuels GΓ’ndoline et PumbΓ© Γ  midi Une famille Γ  connecter, c'est vrai D'un enfant qui voit toi fusionner [Guitar Solo] [Verse 2] Dans le monde des tutos virtuels GΓ’ndoline, PumbΓ© Γ  midi Une famille Γ  connecter, c'est vrai D'un enfant qui voit toi fusionner<|endoftext|> 2026-07-12 10:22:30.900 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:125 - ====================================================================== 2026-07-12 10:22:30.903 | INFO | acestep.core.generation.handler.conditioning_embed:preprocess_batch:110 - [preprocess_batch] Inferring prompt embeddings... 2026-07-12 10:22:30.912 | INFO | acestep.core.generation.handler.conditioning_embed:preprocess_batch:113 - [preprocess_batch] Inferring lyric embeddings... 2026-07-12 10:22:30.912 | INFO | acestep.core.generation.handler.service_generate_execute:_execute_service_generate_diffusion:161 - [service_generate] Generating audio... (DiT backend: PyTorch (cuda)) 2026-07-12 10:22:30.918 | INFO | acestep.core.generation.handler.service_generate_execute:_execute_service_generate_diffusion:263 - [service_generate] DiT diffusion via PyTorch (cuda)... [Request] Loaded request0.json [Turbo] steps=8, shift=3.0 | acestep-v15-turbo-Q5_K_M.gguf [GGML] Running acestep-v15-turbo-Q5_K_M.gguf... [GGML] Done, 47 dump files [Python] Initializing acestep-v15-turbo... [Python] Generating (acestep-v15-turbo, 8 steps)... Using precomputed LM hints Using precomputed LM hints [Python] Wrote python-turbo/output.wav: 4166400 samples (86.80s @ 48kHz stereo) [Python] Done, 40 dump files [Turbo] Cosine similarities GGML vs Python stage GGML vs Python text_hidden 0.999817 lyric_embed 1.000000 enc_hidden 0.999136 detok_output 0.999877 context 0.999921 noise 1.000000 temb_t 0.999972 hidden_after_proj_in 0.999959 enc_after_cond_emb 0.999274 layer0_sa_output 0.999442 hidden_after_layer0 0.999656 hidden_after_layer6 0.999757 hidden_after_layer12 0.998602 hidden_after_layer18 0.995482 hidden_after_layer23 0.991472 dit_step0_vt 0.968016 dit_step0_xt 0.999930 dit_step1_vt 0.970543 dit_step1_xt 0.999779 dit_step2_vt 0.968248 dit_step2_xt 0.999352 dit_step3_vt 0.973090 dit_step3_xt 0.998469 dit_step4_vt 0.971893 dit_step4_xt 0.996493 dit_step5_vt 0.967710 dit_step5_xt 0.991979 dit_step6_vt 0.962905 dit_step6_xt 0.983221 dit_step7_vt 0.954066 dit_x0 0.970056 vae_audio 0.876329 vae_audio (STFT cosine) 0.967723 [Turbo] Error growth GGML vs Python stage cos max_err mean_err mean_A std_A mean_B std_B dit_step0_xt 0.999930 0.143291 0.007836 -0.002280 0.973166 -0.002342 0.972003 dit_step1_xt 0.999779 0.265232 0.013502 -0.005266 0.943146 -0.005313 0.941730 dit_step2_xt 0.999352 0.458384 0.021650 -0.009253 0.909926 -0.009311 0.908527 dit_step3_xt 0.998469 0.667042 0.031998 -0.014605 0.874856 -0.014577 0.873624 dit_step4_xt 0.996493 0.976871 0.046348 -0.021745 0.843313 -0.021660 0.841995 dit_step5_xt 0.991979 1.454744 0.068314 -0.031972 0.827072 -0.032109 0.824593 dit_step6_xt 0.983221 2.238652 0.102044 -0.046361 0.858390 -0.046482 0.855546 [SFT] steps=50, shift=1.0 | acestep-v15-sft-Q5_K_M.gguf [GGML] Running acestep-v15-sft-Q5_K_M.gguf... [GGML] Done, 131 dump files [Python] Initializing acestep-v15-sft... [Python] Generating (acestep-v15-sft, 50 steps)... Using precomputed LM hints Using precomputed LM hints 0%| | 0/50 [00:00 Text-Enc [Registry] Qwen3-Embedding-0.6B-Q8_0.gguf -> Text-Enc [Registry] acestep-5Hz-lm-0.6B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-0.6B-Q8_0.gguf -> LM [Registry] acestep-5Hz-lm-1.7B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-1.7B-Q8_0.gguf -> LM [Registry] acestep-5Hz-lm-4B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q5_K_M.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q6_K.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q8_0.gguf -> LM [Registry] acestep-v15-base-BF16.gguf -> DiT [Registry] acestep-v15-base-Q4_K_M.gguf -> DiT [Registry] acestep-v15-base-Q5_K_M.gguf -> DiT [Registry] acestep-v15-base-Q6_K.gguf -> DiT [Registry] acestep-v15-base-Q8_0.gguf -> DiT [Registry] acestep-v15-sft-BF16.gguf -> DiT [Registry] acestep-v15-sft-Q4_K_M.gguf -> DiT [Registry] acestep-v15-sft-Q5_K_M.gguf -> DiT [Registry] acestep-v15-sft-Q6_K.gguf -> DiT [Registry] acestep-v15-sft-Q8_0.gguf -> DiT [Registry] acestep-v15-sftturbo50-BF16.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q4_K_M.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q5_K_M.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q6_K.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-BF16.gguf -> DiT [Registry] acestep-v15-turbo-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-continuous-BF16.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-shift1-BF16.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-shift3-BF16.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-base-BF16.gguf -> DiT [Registry] acestep-v15-xl-base-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-base-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-base-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-base-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-sft-BF16.gguf -> DiT [Registry] acestep-v15-xl-sft-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-sft-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-sft-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-sft-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-BF16.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-turbo-BF16.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q8_0.gguf -> DiT [Registry] scragvae-BF16.gguf -> VAE [Registry] vae-BF16.gguf -> VAE [Store] Created (policy=STRICT) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-turbo-Q5_K_M.gguf: 830 tensors, data at offset 69088 [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-turbo-Q5_K_M.gguf: 830 tensors, data at offset 69088 [Synth-Load] Ready: turbo=yes, fa=no, batch_cfg=yes [Resolve-Params] max audio codes across batch: 434 (86.8s @ 5Hz) [Resolve-T] T=2170, S=1085 [Resolve-T] seed=42, steps=8, guidance=1.0, shift=3.0, duration=88.0s [BPE] Loaded from GGUF: 151643 vocab, 151387 merges ggml_cuda_init: found 1 CUDA devices (Total VRAM: 97247 MiB): Device 0: NVIDIA RTX PRO 6000 Blackwell Workstation Edition, compute capability 12.0, VMM: yes, VRAM: 97247 MiB load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so ggml_vulkan: Found 1 Vulkan devices: ggml_vulkan: 0 = NVIDIA RTX PRO 6000 Blackwell Workstation Edition (NVIDIA) | uma: 0 | fp16: 1 | bf16: 0 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2 load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] TextEncoder backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/Qwen3-Embedding-0.6B-BF16.gguf: 310 tensors, data at offset 5337568 [Load] TextEncoder: 28L, H=1024, Nh=16/8 [Qwen3] Attn: Q+K+V fused [Qwen3] MLP: gate+up fused [WeightCtx] Loaded 310 tensors, 1136.5 MB into backend [Store] Load TextEnc: 323 ms [Debug] text_hidden: [70, 1024] first4: 3.705836 2.395382 0.221845 -13.145830 [Debug] lyric_embed: [167, 1024] first4: 0.029175 0.032227 -0.022339 -0.028809 [Store] Unload TextEnc (1136.5 MB) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] CondEncoder backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-turbo-Q5_K_M.gguf: 830 tensors, data at offset 69088 [Load] LyricEncoder: 8L [Qwen3] Attn: Q+K fused, V separate [Qwen3] MLP: gate+up fused [Load] TimbreEncoder: 4L [Qwen3] Attn: Q+K fused, V separate [Qwen3] MLP: gate+up fused [WeightCtx] Loaded 141 tensors, 416.2 MB into backend [Load] CondEncoder: lyric(8L), timbre(4L, CLS), text_proj, null_cond [Store] Load CondEnc: 708 ms [CondEnc] Lyric sliding mask: 167x167, window=128 [CondEnc] Timbre sliding mask: 2x2, window=128 (CLS) [Encode] Packed: lyric=167 + timbre=1 + text=70 = 238 tokens [Encode-Text Batch0] 70+167 tokens -> enc_S=238, 8.4 ms [Debug] enc_hidden: [238, 2048] first4: 0.687610 -0.019865 -0.051315 0.022918 [Store] Unload CondEnc (416.2 MB) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] Detokenizer backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-turbo-Q5_K_M.gguf: 830 tensors, data at offset 69088 [WeightCtx] Loaded 30 tensors, 73.2 MB into backend [Load] Detokenizer: FSQ(6->2048) + 2L encoder(S=5, 2048->64) [Store] Load FSQ-Detok: 52 ms [Context] Decoded: 434 codes -> 2170 frames (86.8s @ 25Hz) [Build-Context Batch0] Detokenizer: 170.9 ms, 434 codes [Debug] detok_output: [2170, 64] first4: -0.126205 1.455492 0.291894 -0.650703 [Store] Unload FSQ-Detok (73.2 MB) [Init-Noise Batch0] Philox noise seed=42, [2170, 64] solver=euler [Debug] noise: [2170, 64] first4: 0.194336 2.156250 -0.171875 0.847656 [Debug] context: [2170, 128] first4: -0.126205 1.455492 0.291894 -0.650703 [Init-Noise] Starting: T=2170, S=1085, enc_S=238, steps=8, batch=1 (cover) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] DiT backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-turbo-Q5_K_M.gguf: 830 tensors, data at offset 69088 [DiT] Self-attn: Q+K fused, V separate [DiT] Cross-attn: all separate [DiT] MLP: gate+up fused [Load] null_condition_emb found (CFG available) [WeightCtx] Loaded 630 tensors, 2801.1 MB into backend [Load] DiT: 32 layers, H=2560, Nh=32/8, D=128 [Store] Load DiT: 1932 ms [DiT] Batch N=1, T=2170, S=1085, enc_S=238 [DiT] Graph: 2729 nodes [DiT] Solver: ODE Euler (1 NFE/step, order 1) [Debug] tproj: [15360] first4: 0.008611 0.102928 0.078232 -0.076587 [Debug] temb: [2560] first4: 0.026536 -0.023384 -0.016406 0.047140 [Debug] temb_t: [2560] first4: 0.014694 -0.022190 -0.020725 0.048421 [Debug] temb_r: [2560] first4: 0.011842 -0.001194 0.004319 -0.001281 [Debug] sinusoidal_t: [256] first4: 0.562486 0.789701 0.439822 -0.023583 [Debug] sinusoidal_r: [256] first4: 1.000000 1.000000 1.000000 1.000000 [Debug] temb_lin1_t: [2560] first4: 0.063173 -0.019530 0.138760 -0.115951 [Debug] temb_lin1_r: [2560] first4: -0.008961 0.017832 -0.016914 0.012844 [Debug] hidden_after_proj_in: [2560, 1085] first4: 0.347290 -0.579834 0.248169 -0.175293 [Debug] proj_in_input: [192, 2170] first4: -0.126205 1.455492 0.291894 -0.650703 [Debug] enc_after_cond_emb: [2560, 238] first4: 0.041489 0.656860 0.097504 0.872070 [Debug] layer0_sa_input: [2560, 1085] first4: 0.032966 -0.052398 0.107496 -0.132416 [Debug] layer0_q_after_rope: [128, 32] first4: -0.865154 -2.239372 -2.331339 0.584255 [Debug] layer0_k_after_rope: [128, 8] first4: 1.817383 4.375000 -0.879395 5.601562 [Debug] layer0_sa_output: [2560, 1085] first4: 1.817383 4.375000 -0.879395 5.601562 [Debug] layer0_attn_out: [4096, 1085] first4: -0.865154 -2.239372 -2.331339 0.584255 [Debug] layer0_after_self_attn: [2560, 1085] first4: 0.488784 -2.281324 -0.395761 -2.849367 [Debug] layer0_after_cross_attn: [2560, 1085] first4: 0.339370 -2.003492 -0.270944 -3.555422 [Debug] hidden_after_layer0: [2560, 1085] first4: 1.961321 -2.383363 233.576691 -3.532644 [Debug] hidden_after_layer6: [2560, 1085] first4: -6.225740 -2.331696 310.069092 -12.362862 [Debug] hidden_after_layer12: [2560, 1085] first4: -12.641676 28.906288 252.892853 28.575552 [Debug] hidden_after_layer18: [2560, 1085] first4: 46.526909 17.336054 252.781143 7.227991 [Debug] hidden_after_layer31: [2560, 1085] first4: -115.884071 -200.186356 -159.407745 116.815483 [Debug] dit_step0_vt: [2170, 64] first4: -1.402954 1.749184 0.754913 1.372559 [Debug] dit_step0_xt: [2170, 64] first4: 0.258107 2.076742 -0.206189 0.785267 [DiT] Step 1/8 t=1.000 [Debug] dit_step1_vt: [2170, 64] first4: -1.027954 1.406693 0.854309 1.344482 [Debug] dit_step1_xt: [2170, 64] first4: 0.314177 2.000013 -0.252788 0.711932 [DiT] Step 2/8 t=0.955 [Debug] dit_step2_vt: [2170, 64] first4: -0.681641 1.374481 0.907311 0.920715 [Debug] dit_step2_xt: [2170, 64] first4: 0.359620 1.908381 -0.313275 0.650551 [DiT] Step 3/8 t=0.900 [Debug] dit_step3_vt: [2170, 64] first4: -0.382599 1.413879 0.923355 0.929291 [Debug] dit_step3_xt: [2170, 64] first4: 0.391503 1.790558 -0.390222 0.573110 [DiT] Step 4/8 t=0.833 [Debug] dit_step4_vt: [2170, 64] first4: -0.395355 1.423203 1.044121 0.897312 [Debug] dit_step4_xt: [2170, 64] first4: 0.433862 1.638072 -0.502092 0.476969 [DiT] Step 5/8 t=0.750 [Debug] dit_step5_vt: [2170, 64] first4: 0.006104 2.601379 0.711945 0.723755 [Debug] dit_step5_xt: [2170, 64] first4: 0.432990 1.266446 -0.603798 0.373576 [DiT] Step 6/8 t=0.643 [Debug] dit_step6_vt: [2170, 64] first4: 0.567936 2.369202 0.589577 0.631714 [Debug] dit_step6_xt: [2170, 64] first4: 0.319403 0.792606 -0.721713 0.247233 [DiT] Step 7/8 t=0.500 [Debug] dit_step7_vt: [2170, 64] first4: -1.419678 2.784302 1.331421 0.488892 [Debug] dit_x0: [2170, 64] first4: 0.745306 -0.042685 -1.121140 0.100565 [DiT] Step 8/8 t=0.300 [DiT-Generate] Total: 663.5 ms (663.5 ms/sample) [Debug] dit_output: [2170, 64] first4: 0.745306 -0.042685 -1.121140 0.100565 [Store] Unload DiT (2801.1 MB) [GGUF] /mnt/workspace/git/acestep.cpp/models/vae-BF16.gguf: 365 tensors, data at offset 30048 load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] VAE backend: Vulkan0 (CPU threads: 16) [VAE] Backend: Vulkan0, Weight buffer: 161.1 MB [VAE] Loaded: 5 blocks, upsample=1920x, F32 activations [Store] Load VAE-Dec: 671 ms [VAE] Tiled decode: 3 tiles (chunk=1024, overlap=64, stride=896) [VAE] Graph: 479 nodes, T_latent=960 [VAE] Upsample factor: 1920.00 (expected ~1920) [VAE] Graph: 479 nodes, T_latent=1024 [VAE] Graph: 479 nodes, T_latent=442 [VAE] Tiled decode done: 3 tiles -> T_audio=4166400 (86.80s @ 48kHz) [VAE-Decode Batch0] Decode: 975.1 ms [Debug] vae_audio: [2, 4166400] first4: 0.002549 0.004465 0.004332 0.004129 [Store] Unload VAE-Dec (161.1 MB) [MP3] Encoding 86.8s @ 128 kbps, 48000 Hz stereo [MP3] 1388928 bytes (12.0:1), 455 ms (190.95x realtime), 32 threads [MP3] Wrote ggml-xl-turbo/request00.mp3 [Ace-Synth] All done 2026-07-12 10:22:40.301 | INFO | acestep.core.generation.handler.init_service_loader:_load_main_model_from_checkpoint:174 - [initialize_service] Attempting to load model with attention implementation: sdpa [Python] Wrote python-sft/output.wav: 4166400 samples (86.80s @ 48kHz stereo) [Python] Done, 166 dump files [SFT] Cosine similarities GGML vs Python stage GGML vs Python text_hidden 0.999817 lyric_embed 1.000000 enc_hidden 0.999136 detok_output 0.999877 context 0.999921 noise 1.000000 temb_t 0.999899 hidden_after_proj_in 0.999959 enc_after_cond_emb 0.999272 layer0_sa_output 0.999521 hidden_after_layer0 0.999798 hidden_after_layer6 0.999599 hidden_after_layer12 0.999039 hidden_after_layer18 0.998085 hidden_after_layer23 0.998554 dit_step0_vt 0.998483 dit_step0_xt 1.000000 dit_step5_vt 0.998553 dit_step5_xt 0.999967 dit_step10_vt 0.997879 dit_step10_xt 0.999878 dit_step15_vt 0.996373 dit_step15_xt 0.999679 dit_step20_vt 0.993841 dit_step20_xt 0.999219 dit_step25_vt 0.989117 dit_step25_xt 0.998361 dit_step30_vt 0.983068 dit_step30_xt 0.996905 dit_step35_vt 0.976561 dit_step35_xt 0.994979 dit_step40_vt 0.972268 dit_step40_xt 0.992967 dit_step45_vt 0.976627 dit_step45_xt 0.991499 dit_step49_vt 0.984057 dit_x0 0.990990 vae_audio 0.959625 vae_audio (STFT cosine) 0.983663 [SFT] Error growth GGML vs Python stage cos max_err mean_err mean_A std_A mean_B std_B dit_step0_xt 1.000000 0.017956 0.001568 -0.001635 0.978551 -0.001619 0.978872 dit_step5_xt 0.999967 0.068590 0.005783 -0.008239 0.885331 -0.008480 0.883811 dit_step10_xt 0.999878 0.084370 0.009624 -0.014930 0.801671 -0.015075 0.802324 dit_step15_xt 0.999679 0.139539 0.014153 -0.021610 0.731340 -0.021933 0.731304 dit_step20_xt 0.999219 0.229477 0.020340 -0.028352 0.678627 -0.028882 0.679066 dit_step25_xt 0.998361 0.386174 0.027973 -0.035189 0.647832 -0.035643 0.650780 dit_step30_xt 0.996905 0.602024 0.037828 -0.042063 0.642331 -0.042577 0.647107 dit_step35_xt 0.994979 0.849161 0.049435 -0.048915 0.662997 -0.049471 0.669256 dit_step40_xt 0.992967 1.096595 0.062395 -0.055750 0.707519 -0.056532 0.715783 dit_step45_xt 0.991499 1.319443 0.074734 -0.062634 0.771592 -0.063533 0.780575 [Xl-turbo] steps=8, shift=3.0 | acestep-v15-xl-turbo-Q5_K_M.gguf [GGML] Running acestep-v15-xl-turbo-Q5_K_M.gguf... [GGML] Done, 47 dump files [Python] Initializing acestep-v15-xl-turbo... Loading checkpoint shards: 0%| | 0/4 [00:00 1.0 (turbo does not use CFG). 2026-07-12 10:22:44.004 | INFO | acestep.core.generation.handler.generate_music:generate_music:304 - [generate_music] Starting generation... 2026-07-12 10:22:44.004 | INFO | acestep.core.generation.handler.generate_music:generate_music:307 - [generate_music] Preparing inputs... 2026-07-12 10:22:44.004 | INFO | acestep.core.generation.handler.generate_music_request:_prepare_reference_and_source_audio:179 - [generate_music] cover task: no src_audio but audio codes provided, proceeding with codes 2026-07-12 10:22:44.007 | INFO | acestep.core.generation.handler.generate_music:_vram_preflight_check:156 - [generate_music] VRAM pre-flight: 76.75 GB free, ~1.23 GB needed (batch=1, duration=88s, mode=xl_turbo). 2026-07-12 10:22:44.012 | INFO | acestep.core.generation.handler.conditioning_target:_prepare_target_latents_and_wavs:78 - [generate_music] Decoding audio codes for item 0... 2026-07-12 10:22:44.018 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_precomputed_lm_hints:31 - [generate_music] Decoding audio codes for LM hints for item 0... 2026-07-12 10:22:44.020 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:119 - ====================================================================== 2026-07-12 10:22:44.020 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:120 - πŸ” [DEBUG] DiT TEXT ENCODER INPUT (Inference) 2026-07-12 10:22:44.020 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:121 - ====================================================================== 2026-07-12 10:22:44.020 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:122 - text_prompt: # Instruction Generate audio semantic tokens based on the given conditions: # Caption An upbeat and anthemic pop-rock track driven by bright, slightly overdriven # Metas - bpm: 83 - timesignature: 4 - keyscale: G major - duration: 88 seconds <|endoftext|> 2026-07-12 10:22:44.020 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:123 - ====================================================================== 2026-07-12 10:22:44.020 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:124 - lyrics_text: # Languages fr # Lyric # Lyric [Intro - Guitar Riff] [Verse 1] Dans le monde des tutos virtuels G ta toise en nouvelle passion Avec Ggendoline et PumbΓ© Γ  midi La communautΓ©, c'est l'unitΓ© Quel joie, une clΓ© [Chorus] Dans le monde des tutos virtuels GΓ’ndoline et PumbΓ© Γ  midi Une famille Γ  connecter, c'est vrai D'un enfant qui voit toi fusionner [Guitar Solo] [Verse 2] Dans le monde des tutos virtuels GΓ’ndoline, PumbΓ© Γ  midi Une famille Γ  connecter, c'est vrai D'un enfant qui voit toi fusionner<|endoftext|> 2026-07-12 10:22:44.020 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:125 - ====================================================================== 2026-07-12 10:22:44.024 | INFO | acestep.core.generation.handler.conditioning_embed:preprocess_batch:110 - [preprocess_batch] Inferring prompt embeddings... 2026-07-12 10:22:44.031 | INFO | acestep.core.generation.handler.conditioning_embed:preprocess_batch:113 - [preprocess_batch] Inferring lyric embeddings... 2026-07-12 10:22:44.031 | INFO | acestep.core.generation.handler.service_generate_execute:_execute_service_generate_diffusion:161 - [service_generate] Generating audio... (DiT backend: PyTorch (cuda)) 2026-07-12 10:22:44.038 | INFO | acestep.core.generation.handler.service_generate_execute:_execute_service_generate_diffusion:263 - [service_generate] DiT diffusion via PyTorch (cuda)... 2026-07-12 10:22:44.044 | INFO | acestep.models.common.dcw_correction:__init__:98 - [DCW] Active β€” mode=double, scaler=0.0500, high_scaler=0.0200, wavelet='haar' 2026-07-12 10:22:44.601 | INFO | acestep.core.generation.handler.generate_music_decode:_prepare_generate_music_decode_state:42 - [generate_music] Model generation completed. Decoding latents... 2026-07-12 10:22:44.611 | DEBUG | acestep.core.generation.handler.generate_music_decode:_prepare_generate_music_decode_state:64 - [generate_music] pred_latents: torch.Size([1, 2170, 64]), dtype=torch.bfloat16 2026-07-12 10:22:44.611 | DEBUG | acestep.core.generation.handler.generate_music_decode:_prepare_generate_music_decode_state:65 - [generate_music] time_costs: {'encoder_time_cost': 0.006582498550415039, 'diffusion_time_cost': 0.5560781955718994, 'diffusion_per_step_time_cost': 0.06950977444648743, 'total_time_cost': 0.5626606941223145, 'offload_time_cost': 0.0} 2026-07-12 10:22:44.625 | INFO | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:127 - [generate_music] Decoding latents with VAE... 2026-07-12 10:22:44.628 | DEBUG | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:136 - [generate_music] Before VAE decode: allocated=17.03GB, max=18.34GB 2026-07-12 10:22:44.628 | INFO | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:154 - [generate_music] Effective free VRAM before VAE decode: 76.66 GB 2026-07-12 10:22:44.628 | INFO | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:172 - [generate_music] Using tiled VAE decode to reduce VRAM usage... 2026-07-12 10:22:44.628 | DEBUG | acestep.core.generation.handler.memory_utils:_get_auto_decode_chunk_size:108 - [_get_auto_decode_chunk_size] Effective free VRAM: 76.66 GB 2026-07-12 10:22:44.628 | DEBUG | acestep.core.generation.handler.memory_utils:_should_offload_wav_to_cpu:131 - [_should_offload_wav_to_cpu] Effective free VRAM: 76.66 GB 2026-07-12 10:22:44.628 | INFO | acestep.core.generation.handler.vae_decode:tiled_decode:56 - [tiled_decode] chunk_size=512, offload_wav_to_cpu=False, latents_shape=torch.Size([1, 64, 2170]) 2026-07-12 10:22:44.904 | DEBUG | acestep.core.generation.handler.generate_music_decode:_decode_generate_music_pred_latents:193 - [generate_music] After VAE decode: allocated=17.21GB, max=18.50GB 2026-07-12 10:22:45.112 | INFO | acestep.core.generation.handler.generate_music_payload:_build_generate_music_success_payload:38 - [generate_music] VAE decode completed. Preparing audio tensors... 2026-07-12 10:22:45.116 | INFO | acestep.core.generation.handler.generate_music_payload:_build_generate_music_success_payload:50 - [generate_music] Done! Generated 1 audio tensors. [Request] Parsed ggml-xl-sft/request0.json [Request] seed=42 lm_batch=1 synth_batch=1 [Request] caption: An upbeat and anthemic pop-rock track driven by bright, slig... [Request] lyrics: 490 bytes [Request] bpm=83 dur=88 key=G major ts=4 lang=fr [Request] lm: temp=0.85 cfg=2.0 top_p=0.90 top_k=0 [Request] dit: steps=50 guidance=1.0 shift=1.0 [Request] task_type: text2music [Request] solver: euler (stork_substeps=10) [Request] lm_mode: generate [Request] output_format: mp3 [Request] synth_model: acestep-v15-xl-sft-Q5_K_M.gguf [Request] vae: vae-BF16.gguf [Request] audio_codes: (present) [Ace-Synth] Batch: 1 request(s) [Registry] Qwen3-Embedding-0.6B-BF16.gguf -> Text-Enc [Registry] Qwen3-Embedding-0.6B-Q8_0.gguf -> Text-Enc [Registry] acestep-5Hz-lm-0.6B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-0.6B-Q8_0.gguf -> LM [Registry] acestep-5Hz-lm-1.7B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-1.7B-Q8_0.gguf -> LM [Registry] acestep-5Hz-lm-4B-BF16.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q5_K_M.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q6_K.gguf -> LM [Registry] acestep-5Hz-lm-4B-Q8_0.gguf -> LM [Registry] acestep-v15-base-BF16.gguf -> DiT [Registry] acestep-v15-base-Q4_K_M.gguf -> DiT [Registry] acestep-v15-base-Q5_K_M.gguf -> DiT [Registry] acestep-v15-base-Q6_K.gguf -> DiT [Registry] acestep-v15-base-Q8_0.gguf -> DiT [Registry] acestep-v15-sft-BF16.gguf -> DiT [Registry] acestep-v15-sft-Q4_K_M.gguf -> DiT [Registry] acestep-v15-sft-Q5_K_M.gguf -> DiT [Registry] acestep-v15-sft-Q6_K.gguf -> DiT [Registry] acestep-v15-sft-Q8_0.gguf -> DiT [Registry] acestep-v15-sftturbo50-BF16.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q4_K_M.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q5_K_M.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q6_K.gguf -> DiT [Registry] acestep-v15-sftturbo50-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-BF16.gguf -> DiT [Registry] acestep-v15-turbo-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-continuous-BF16.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-continuous-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-shift1-BF16.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-shift1-Q8_0.gguf -> DiT [Registry] acestep-v15-turbo-shift3-BF16.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q4_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q5_K_M.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q6_K.gguf -> DiT [Registry] acestep-v15-turbo-shift3-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-base-BF16.gguf -> DiT [Registry] acestep-v15-xl-base-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-base-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-base-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-base-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-sft-BF16.gguf -> DiT [Registry] acestep-v15-xl-sft-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-sft-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-sft-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-sft-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-BF16.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-sftturbo50-Q8_0.gguf -> DiT [Registry] acestep-v15-xl-turbo-BF16.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q4_K_M.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q5_K_M.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q6_K.gguf -> DiT [Registry] acestep-v15-xl-turbo-Q8_0.gguf -> DiT [Registry] scragvae-BF16.gguf -> VAE [Registry] vae-BF16.gguf -> VAE [Store] Created (policy=STRICT) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-sft-Q5_K_M.gguf: 830 tensors, data at offset 69056 [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-sft-Q5_K_M.gguf: 830 tensors, data at offset 69056 [Synth-Load] Ready: turbo=no, fa=no, batch_cfg=yes [Resolve-Params] max audio codes across batch: 434 (86.8s @ 5Hz) [Resolve-T] T=2170, S=1085 [Resolve-T] seed=42, steps=50, guidance=1.0, shift=1.0, duration=88.0s [BPE] Loaded from GGUF: 151643 vocab, 151387 merges ggml_cuda_init: found 1 CUDA devices (Total VRAM: 97247 MiB): Device 0: NVIDIA RTX PRO 6000 Blackwell Workstation Edition, compute capability 12.0, VMM: yes, VRAM: 97247 MiB load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so ggml_vulkan: Found 1 Vulkan devices: ggml_vulkan: 0 = NVIDIA RTX PRO 6000 Blackwell Workstation Edition (NVIDIA) | uma: 0 | fp16: 1 | bf16: 0 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2 load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] TextEncoder backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/Qwen3-Embedding-0.6B-BF16.gguf: 310 tensors, data at offset 5337568 [Load] TextEncoder: 28L, H=1024, Nh=16/8 [Qwen3] Attn: Q+K+V fused [Qwen3] MLP: gate+up fused [WeightCtx] Loaded 310 tensors, 1136.5 MB into backend [Store] Load TextEnc: 512 ms [Debug] text_hidden: [70, 1024] first4: 3.705836 2.395382 0.221845 -13.145830 [Debug] lyric_embed: [167, 1024] first4: 0.029175 0.032227 -0.022339 -0.028809 [Store] Unload TextEnc (1136.5 MB) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] CondEncoder backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-sft-Q5_K_M.gguf: 830 tensors, data at offset 69056 [Load] LyricEncoder: 8L [Qwen3] Attn: Q+K fused, V separate [Qwen3] MLP: gate+up fused [Load] TimbreEncoder: 4L [Qwen3] Attn: Q+K fused, V separate [Qwen3] MLP: gate+up fused [WeightCtx] Loaded 141 tensors, 416.2 MB into backend [Load] CondEncoder: lyric(8L), timbre(4L, CLS), text_proj, null_cond [Store] Load CondEnc: 737 ms [CondEnc] Lyric sliding mask: 167x167, window=128 [CondEnc] Timbre sliding mask: 2x2, window=128 (CLS) [Encode] Packed: lyric=167 + timbre=1 + text=70 = 238 tokens [Encode-Text Batch0] 70+167 tokens -> enc_S=238, 19.7 ms [Debug] enc_hidden: [238, 2048] first4: 0.687610 -0.019865 -0.051315 0.022918 [Store] Unload CondEnc (416.2 MB) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] Detokenizer backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-sft-Q5_K_M.gguf: 830 tensors, data at offset 69056 [WeightCtx] Loaded 30 tensors, 73.2 MB into backend [Load] Detokenizer: FSQ(6->2048) + 2L encoder(S=5, 2048->64) [Store] Load FSQ-Detok: 53 ms [Context] Decoded: 434 codes -> 2170 frames (86.8s @ 25Hz) [Build-Context Batch0] Detokenizer: 166.6 ms, 434 codes [Debug] detok_output: [2170, 64] first4: -0.126205 1.455492 0.291894 -0.650703 [Store] Unload FSQ-Detok (73.2 MB) [Init-Noise Batch0] Philox noise seed=42, [2170, 64] solver=euler [Debug] noise: [2170, 64] first4: 0.194336 2.156250 -0.171875 0.847656 [Debug] context: [2170, 128] first4: -0.126205 1.455492 0.291894 -0.650703 [Init-Noise] Starting: T=2170, S=1085, enc_S=238, steps=50, batch=1 (cover) load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] DiT backend: Vulkan0 (CPU threads: 16) [GGUF] /mnt/workspace/git/acestep.cpp/models/acestep-v15-xl-sft-Q5_K_M.gguf: 830 tensors, data at offset 69056 [DiT] Self-attn: Q+K fused, V separate [DiT] Cross-attn: all separate [DiT] MLP: gate+up fused [Load] null_condition_emb found (CFG available) [WeightCtx] Loaded 630 tensors, 2801.1 MB into backend [Load] DiT: 32 layers, H=2560, Nh=32/8, D=128 [Store] Load DiT: 1950 ms [DiT] Batch N=1, T=2170, S=1085, enc_S=238 [DiT] Graph: 2729 nodes [DiT] Solver: ODE Euler (1 NFE/step, order 1) [Debug] tproj: [15360] first4: 0.005156 0.089698 0.077651 -0.035042 [Debug] temb: [2560] first4: 0.009955 -0.008020 -0.006084 0.004950 [Debug] temb_t: [2560] first4: 0.008419 -0.008148 -0.006723 0.003948 [Debug] temb_r: [2560] first4: 0.001536 0.000128 0.000639 0.001002 [Debug] sinusoidal_t: [256] first4: 0.562486 0.789701 0.439822 -0.023583 [Debug] sinusoidal_r: [256] first4: 1.000000 1.000000 1.000000 1.000000 [Debug] temb_lin1_t: [2560] first4: 0.028513 -0.079992 0.086633 -0.070856 [Debug] temb_lin1_r: [2560] first4: 0.006506 -0.006323 -0.000918 -0.013460 [Debug] hidden_after_proj_in: [2560, 1085] first4: 0.337036 -0.495361 0.120361 -0.198975 [Debug] proj_in_input: [192, 2170] first4: -0.126205 1.455492 0.291894 -0.650703 [Debug] enc_after_cond_emb: [2560, 238] first4: 0.052849 0.649338 0.105530 0.866615 [Debug] layer0_sa_input: [2560, 1085] first4: 0.029791 -0.035540 0.103631 -0.096009 [Debug] layer0_q_after_rope: [128, 32] first4: -0.867917 -2.225467 -2.790677 0.573406 [Debug] layer0_k_after_rope: [128, 8] first4: 2.195312 4.488281 -1.233398 8.062500 [Debug] layer0_sa_output: [2560, 1085] first4: 2.195312 4.488281 -1.233398 8.062500 [Debug] layer0_attn_out: [4096, 1085] first4: -0.867917 -2.225467 -2.790677 0.573406 [Debug] layer0_after_self_attn: [2560, 1085] first4: 0.474917 -2.224739 -0.792218 -4.254282 [Debug] layer0_after_cross_attn: [2560, 1085] first4: 0.284976 -1.939583 -0.732403 -4.281702 [Debug] hidden_after_layer0: [2560, 1085] first4: -1.412108 -1.285975 246.193710 -4.118339 [Debug] hidden_after_layer6: [2560, 1085] first4: -7.606342 2.999075 303.915985 -20.950941 [Debug] hidden_after_layer12: [2560, 1085] first4: -21.434332 38.087337 219.991257 16.968704 [Debug] hidden_after_layer18: [2560, 1085] first4: 11.757919 -15.202559 228.412186 19.643681 [Debug] hidden_after_layer31: [2560, 1085] first4: 67.997841 -113.903381 -1.389866 86.181534 [Debug] dit_step0_vt: [2170, 64] first4: -0.340942 2.257080 0.037354 1.709045 [Debug] dit_step0_xt: [2170, 64] first4: 0.201155 2.111109 -0.172622 0.813475 [DiT] Step 1/50 t=1.000 [Debug] dit_step1_vt: [2170, 64] first4: -0.369995 2.194458 0.101562 1.562439 [Debug] dit_step1_xt: [2170, 64] first4: 0.208555 2.067219 -0.174653 0.782227 [DiT] Step 2/50 t=0.980 [Debug] dit_step2_vt: [2170, 64] first4: -0.353638 2.203979 0.107422 1.552734 [Debug] dit_step2_xt: [2170, 64] first4: 0.215627 2.023140 -0.176802 0.751172 [DiT] Step 3/50 t=0.960 [Debug] dit_step3_vt: [2170, 64] first4: -0.305664 2.119446 0.121460 1.548340 [Debug] dit_step3_xt: [2170, 64] first4: 0.221741 1.980751 -0.179231 0.720205 [DiT] Step 4/50 t=0.940 [Debug] dit_step4_vt: [2170, 64] first4: -0.305176 2.055542 0.098145 1.570839 [Debug] dit_step4_xt: [2170, 64] first4: 0.227844 1.939640 -0.181194 0.688788 [DiT] Step 5/50 t=0.920 [Debug] dit_step5_vt: [2170, 64] first4: -0.297729 2.032196 0.108276 1.576385 [Debug] dit_step5_xt: [2170, 64] first4: 0.233799 1.898996 -0.183359 0.657261 [DiT] Step 6/50 t=0.900 [Debug] dit_step6_vt: [2170, 64] first4: -0.293945 2.009460 0.129639 1.596985 [Debug] dit_step6_xt: [2170, 64] first4: 0.239678 1.858807 -0.185952 0.625321 [DiT] Step 7/50 t=0.880 [Debug] dit_step7_vt: [2170, 64] first4: -0.296875 1.971069 0.157837 1.599030 [Debug] dit_step7_xt: [2170, 64] first4: 0.245615 1.819386 -0.189109 0.593340 [DiT] Step 8/50 t=0.860 [Debug] dit_step8_vt: [2170, 64] first4: -0.313721 1.926453 0.187500 1.605469 [Debug] dit_step8_xt: [2170, 64] first4: 0.251890 1.780856 -0.192859 0.561231 [DiT] Step 9/50 t=0.840 [Debug] dit_step9_vt: [2170, 64] first4: -0.340698 1.864594 0.160400 1.578926 [Debug] dit_step9_xt: [2170, 64] first4: 0.258704 1.743564 -0.196067 0.529652 [DiT] Step 10/50 t=0.820 [Debug] dit_step10_vt: [2170, 64] first4: -0.352539 1.826172 0.143799 1.572792 [Debug] dit_step10_xt: [2170, 64] first4: 0.265754 1.707041 -0.198943 0.498196 [DiT] Step 11/50 t=0.800 [Debug] dit_step11_vt: [2170, 64] first4: -0.346191 1.784042 0.129028 1.561111 [Debug] dit_step11_xt: [2170, 64] first4: 0.272678 1.671360 -0.201523 0.466974 [DiT] Step 12/50 t=0.780 [Debug] dit_step12_vt: [2170, 64] first4: -0.344971 1.731033 0.119507 1.546051 [Debug] dit_step12_xt: [2170, 64] first4: 0.279578 1.636740 -0.203914 0.436053 [DiT] Step 13/50 t=0.760 [Debug] dit_step13_vt: [2170, 64] first4: -0.327271 1.680416 0.094604 1.532654 [Debug] dit_step13_xt: [2170, 64] first4: 0.286123 1.603131 -0.205806 0.405400 [DiT] Step 14/50 t=0.740 [Debug] dit_step14_vt: [2170, 64] first4: -0.318237 1.628029 0.076904 1.522919 [Debug] dit_step14_xt: [2170, 64] first4: 0.292488 1.570571 -0.207344 0.374942 [DiT] Step 15/50 t=0.720 [Debug] dit_step15_vt: [2170, 64] first4: -0.303955 1.578690 0.052612 1.505737 [Debug] dit_step15_xt: [2170, 64] first4: 0.298567 1.538997 -0.208396 0.344827 [DiT] Step 16/50 t=0.700 [Debug] dit_step16_vt: [2170, 64] first4: -0.268677 1.528992 0.027588 1.499512 [Debug] dit_step16_xt: [2170, 64] first4: 0.303940 1.508417 -0.208948 0.314837 [DiT] Step 17/50 t=0.680 [Debug] dit_step17_vt: [2170, 64] first4: -0.239258 1.479370 0.011108 1.487122 [Debug] dit_step17_xt: [2170, 64] first4: 0.308726 1.478830 -0.209170 0.285094 [DiT] Step 18/50 t=0.660 [Debug] dit_step18_vt: [2170, 64] first4: -0.194397 1.424255 -0.002686 1.487366 [Debug] dit_step18_xt: [2170, 64] first4: 0.312613 1.450345 -0.209116 0.255347 [DiT] Step 19/50 t=0.640 [Debug] dit_step19_vt: [2170, 64] first4: -0.168823 1.371338 -0.014648 1.479919 [Debug] dit_step19_xt: [2170, 64] first4: 0.315990 1.422918 -0.208823 0.225749 [DiT] Step 20/50 t=0.620 [Debug] dit_step20_vt: [2170, 64] first4: -0.128296 1.331299 -0.018555 1.472015 [Debug] dit_step20_xt: [2170, 64] first4: 0.318556 1.396292 -0.208452 0.196308 [DiT] Step 21/50 t=0.600 [Debug] dit_step21_vt: [2170, 64] first4: -0.098328 1.280090 -0.029663 1.455750 [Debug] dit_step21_xt: [2170, 64] first4: 0.320522 1.370690 -0.207859 0.167193 [DiT] Step 22/50 t=0.580 [Debug] dit_step22_vt: [2170, 64] first4: -0.072205 1.234924 -0.030640 1.436462 [Debug] dit_step22_xt: [2170, 64] first4: 0.321967 1.345991 -0.207246 0.138464 [DiT] Step 23/50 t=0.560 [Debug] dit_step23_vt: [2170, 64] first4: -0.048340 1.183350 -0.043945 1.405518 [Debug] dit_step23_xt: [2170, 64] first4: 0.322933 1.322325 -0.206367 0.110354 [DiT] Step 24/50 t=0.540 [Debug] dit_step24_vt: [2170, 64] first4: -0.016632 1.135742 -0.047241 1.398193 [Debug] dit_step24_xt: [2170, 64] first4: 0.323266 1.299610 -0.205422 0.082390 [DiT] Step 25/50 t=0.520 [Debug] dit_step25_vt: [2170, 64] first4: -0.009827 1.077454 -0.058228 1.373657 [Debug] dit_step25_xt: [2170, 64] first4: 0.323463 1.278061 -0.204258 0.054917 [DiT] Step 26/50 t=0.500 [Debug] dit_step26_vt: [2170, 64] first4: -0.004974 1.019318 -0.068970 1.352844 [Debug] dit_step26_xt: [2170, 64] first4: 0.323562 1.257674 -0.202878 0.027860 [DiT] Step 27/50 t=0.480 [Debug] dit_step27_vt: [2170, 64] first4: -0.011780 0.953918 -0.083557 1.326233 [Debug] dit_step27_xt: [2170, 64] first4: 0.323798 1.238596 -0.201207 0.001335 [DiT] Step 28/50 t=0.460 [Debug] dit_step28_vt: [2170, 64] first4: -0.016907 0.896881 -0.090149 1.302551 [Debug] dit_step28_xt: [2170, 64] first4: 0.324136 1.220658 -0.199404 -0.024716 [DiT] Step 29/50 t=0.440 [Debug] dit_step29_vt: [2170, 64] first4: -0.017334 0.826935 -0.101562 1.280518 [Debug] dit_step29_xt: [2170, 64] first4: 0.324482 1.204119 -0.197373 -0.050326 [DiT] Step 30/50 t=0.420 [Debug] dit_step30_vt: [2170, 64] first4: -0.024536 0.752205 -0.121277 1.270782 [Debug] dit_step30_xt: [2170, 64] first4: 0.324973 1.189075 -0.194948 -0.075742 [DiT] Step 31/50 t=0.400 [Debug] dit_step31_vt: [2170, 64] first4: -0.039368 0.683868 -0.102112 1.256195 [Debug] dit_step31_xt: [2170, 64] first4: 0.325760 1.175398 -0.192905 -0.100866 [DiT] Step 32/50 t=0.380 [Debug] dit_step32_vt: [2170, 64] first4: -0.044556 0.603951 -0.105591 1.210724 [Debug] dit_step32_xt: [2170, 64] first4: 0.326652 1.163319 -0.190793 -0.125080 [DiT] Step 33/50 t=0.360 [Debug] dit_step33_vt: [2170, 64] first4: -0.047241 0.517654 -0.110748 1.179749 [Debug] dit_step33_xt: [2170, 64] first4: 0.327596 1.152966 -0.188579 -0.148675 [DiT] Step 34/50 t=0.340 [Debug] dit_step34_vt: [2170, 64] first4: -0.058807 0.423027 -0.119965 1.156281 [Debug] dit_step34_xt: [2170, 64] first4: 0.328773 1.144505 -0.186179 -0.171801 [DiT] Step 35/50 t=0.320 [Debug] dit_step35_vt: [2170, 64] first4: -0.076080 0.320877 -0.134399 1.124054 [Debug] dit_step35_xt: [2170, 64] first4: 0.330294 1.138088 -0.183491 -0.194282 [DiT] Step 36/50 t=0.300 [Debug] dit_step36_vt: [2170, 64] first4: -0.103653 0.221161 -0.126396 1.087738 [Debug] dit_step36_xt: [2170, 64] first4: 0.332367 1.133665 -0.180963 -0.216037 [DiT] Step 37/50 t=0.280 [Debug] dit_step37_vt: [2170, 64] first4: -0.113057 0.122864 -0.133240 1.057129 [Debug] dit_step37_xt: [2170, 64] first4: 0.334628 1.131207 -0.178299 -0.237179 [DiT] Step 38/50 t=0.260 [Debug] dit_step38_vt: [2170, 64] first4: -0.140434 0.021484 -0.099426 1.006836 [Debug] dit_step38_xt: [2170, 64] first4: 0.337437 1.130778 -0.176310 -0.257316 [DiT] Step 39/50 t=0.240 [Debug] dit_step39_vt: [2170, 64] first4: -0.164139 -0.085052 -0.077042 0.986328 [Debug] dit_step39_xt: [2170, 64] first4: 0.340720 1.132479 -0.174769 -0.277042 [DiT] Step 40/50 t=0.220 [Debug] dit_step40_vt: [2170, 64] first4: -0.200043 -0.196350 -0.075439 0.927429 [Debug] dit_step40_xt: [2170, 64] first4: 0.344721 1.136406 -0.173260 -0.295591 [DiT] Step 41/50 t=0.200 [Debug] dit_step41_vt: [2170, 64] first4: -0.224548 -0.295536 -0.051453 0.886353 [Debug] dit_step41_xt: [2170, 64] first4: 0.349212 1.142316 -0.172231 -0.313318 [DiT] Step 42/50 t=0.180 [Debug] dit_step42_vt: [2170, 64] first4: -0.244690 -0.380402 -0.018616 0.832397 [Debug] dit_step42_xt: [2170, 64] first4: 0.354105 1.149925 -0.171859 -0.329966 [DiT] Step 43/50 t=0.160 [Debug] dit_step43_vt: [2170, 64] first4: -0.277527 -0.478165 -0.002575 0.786926 [Debug] dit_step43_xt: [2170, 64] first4: 0.359656 1.159488 -0.171807 -0.345705 [DiT] Step 44/50 t=0.140 [Debug] dit_step44_vt: [2170, 64] first4: -0.293915 -0.573761 0.056976 0.724426 [Debug] dit_step44_xt: [2170, 64] first4: 0.365534 1.170963 -0.172947 -0.360193 [DiT] Step 45/50 t=0.120 [Debug] dit_step45_vt: [2170, 64] first4: -0.320145 -0.658112 0.087677 0.656372 [Debug] dit_step45_xt: [2170, 64] first4: 0.371937 1.184125 -0.174701 -0.373321 [DiT] Step 46/50 t=0.100 [Debug] dit_step46_vt: [2170, 64] first4: -0.358597 -0.725754 0.123596 0.614807 [Debug] dit_step46_xt: [2170, 64] first4: 0.379109 1.198640 -0.177172 -0.385617 [DiT] Step 47/50 t=0.080 [Debug] dit_step47_vt: [2170, 64] first4: -0.404517 -0.798599 0.191772 0.581787 [Debug] dit_step47_xt: [2170, 64] first4: 0.387199 1.214612 -0.181008 -0.397252 [DiT] Step 48/50 t=0.060 [Debug] dit_step48_vt: [2170, 64] first4: -0.454727 -0.861923 0.247070 0.568970 [Debug] dit_step48_xt: [2170, 64] first4: 0.396294 1.231851 -0.185949 -0.408632 [DiT] Step 49/50 t=0.040 [Debug] dit_step49_vt: [2170, 64] first4: -0.445557 -0.842712 0.171814 0.580383 [Debug] dit_x0: [2170, 64] first4: 0.405205 1.248705 -0.189386 -0.420240 [DiT] Step 50/50 t=0.020 [DiT-Generate] Total: 3931.4 ms (3931.4 ms/sample) [Debug] dit_output: [2170, 64] first4: 0.405205 1.248705 -0.189386 -0.420240 [Store] Unload DiT (2801.1 MB) [GGUF] /mnt/workspace/git/acestep.cpp/models/vae-BF16.gguf: 365 tensors, data at offset 30048 load_backend: loaded CUDA backend from /mnt/workspace/git/acestep.cpp/build/libggml-cuda.so load_backend: loaded Vulkan backend from /mnt/workspace/git/acestep.cpp/build/libggml-vulkan.so load_backend: loaded CPU backend from /mnt/workspace/git/acestep.cpp/build/libggml-cpu-zen4.so [Load] VAE backend: Vulkan0 (CPU threads: 16) [VAE] Backend: Vulkan0, Weight buffer: 161.1 MB [VAE] Loaded: 5 blocks, upsample=1920x, F32 activations [Store] Load VAE-Dec: 676 ms [VAE] Tiled decode: 3 tiles (chunk=1024, overlap=64, stride=896) [VAE] Graph: 479 nodes, T_latent=960 [VAE] Upsample factor: 1920.00 (expected ~1920) [VAE] Graph: 479 nodes, T_latent=1024 [VAE] Graph: 479 nodes, T_latent=442 [VAE] Tiled decode done: 3 tiles -> T_audio=4166400 (86.80s @ 48kHz) [VAE-Decode Batch0] Decode: 967.0 ms [Debug] vae_audio: [2, 4166400] first4: -0.001553 -0.000841 -0.000953 -0.000432 [Store] Unload VAE-Dec (161.1 MB) [MP3] Encoding 86.8s @ 128 kbps, 48000 Hz stereo [MP3] 1388928 bytes (12.0:1), 404 ms (214.81x realtime), 32 threads [MP3] Wrote ggml-xl-sft/request00.mp3 [Ace-Synth] All done 2026-07-12 10:22:55.844 | INFO | acestep.core.generation.handler.init_service_loader:_load_main_model_from_checkpoint:174 - [initialize_service] Attempting to load model with attention implementation: sdpa [Python] Generating (acestep-v15-xl-turbo, 8 steps)... Using precomputed LM hints Using precomputed LM hints [Python] Wrote python-xl-turbo/output.wav: 4166400 samples (86.80s @ 48kHz stereo) [Python] Done, 40 dump files [Xl-turbo] Cosine similarities GGML vs Python stage GGML vs Python text_hidden 0.999817 lyric_embed 1.000000 enc_hidden 0.998898 detok_output 0.999877 context 0.999921 noise 1.000000 temb_t 0.999907 hidden_after_proj_in 0.999940 enc_after_cond_emb 0.998876 layer0_sa_output 0.988768 hidden_after_layer0 0.999256 hidden_after_layer6 0.999609 hidden_after_layer12 0.994648 hidden_after_layer18 0.995441 hidden_after_layer31 0.987338 dit_step0_vt 0.908322 dit_step0_xt 0.999835 dit_step1_vt 0.921729 dit_step1_xt 0.999485 dit_step2_vt 0.918194 dit_step2_xt 0.998621 dit_step3_vt 0.921507 dit_step3_xt 0.996723 dit_step4_vt 0.913350 dit_step4_xt 0.991951 dit_step5_vt 0.905095 dit_step5_xt 0.980544 dit_step6_vt 0.895637 dit_step6_xt 0.957691 dit_step7_vt 0.872850 dit_x0 0.920173 vae_audio 0.726927 vae_audio (STFT cosine) 0.918576 [Xl-turbo] Error growth GGML vs Python stage cos max_err mean_err mean_A std_A mean_B std_B dit_step0_xt 0.999835 0.232472 0.012944 -0.002876 0.978387 -0.002721 0.976534 dit_step1_xt 0.999485 0.276700 0.022594 -0.005740 0.952705 -0.005567 0.949004 dit_step2_xt 0.998621 0.460397 0.035680 -0.009348 0.924334 -0.009330 0.918931 dit_step3_xt 0.996723 0.775736 0.052306 -0.014900 0.893896 -0.014418 0.887241 dit_step4_xt 0.991951 1.370384 0.078335 -0.022583 0.865803 -0.021453 0.857796 dit_step5_xt 0.980544 2.039242 0.118686 -0.033991 0.849486 -0.032240 0.840033 dit_step6_xt 0.957691 2.862118 0.180557 -0.050890 0.874231 -0.049194 0.863964 [Xl-sft] steps=50, shift=1.0 | acestep-v15-xl-sft-Q5_K_M.gguf [GGML] Running acestep-v15-xl-sft-Q5_K_M.gguf... [GGML] Done, 131 dump files [Python] Initializing acestep-v15-xl-sft... Loading checkpoint shards: 0%| | 0/4 [00:00 2026-07-12 10:23:01.427 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:123 - ====================================================================== 2026-07-12 10:23:01.427 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:124 - lyrics_text: # Languages fr # Lyric # Lyric [Intro - Guitar Riff] [Verse 1] Dans le monde des tutos virtuels G ta toise en nouvelle passion Avec Ggendoline et PumbΓ© Γ  midi La communautΓ©, c'est l'unitΓ© Quel joie, une clΓ© [Chorus] Dans le monde des tutos virtuels GΓ’ndoline et PumbΓ© Γ  midi Une famille Γ  connecter, c'est vrai D'un enfant qui voit toi fusionner [Guitar Solo] [Verse 2] Dans le monde des tutos virtuels GΓ’ndoline, PumbΓ© Γ  midi Une famille Γ  connecter, c'est vrai D'un enfant qui voit toi fusionner<|endoftext|> 2026-07-12 10:23:01.427 | INFO | acestep.core.generation.handler.conditioning_text:_prepare_text_conditioning_inputs:125 - ====================================================================== 2026-07-12 10:23:01.431 | INFO | acestep.core.generation.handler.conditioning_embed:preprocess_batch:110 - [preprocess_batch] Inferring prompt embeddings... 2026-07-12 10:23:01.438 | INFO | acestep.core.generation.handler.conditioning_embed:preprocess_batch:113 - [preprocess_batch] Inferring lyric embeddings... 2026-07-12 10:23:01.438 | INFO | acestep.core.generation.handler.service_generate_execute:_execute_service_generate_diffusion:161 - [service_generate] Generating audio... (DiT backend: PyTorch (cuda)) 2026-07-12 10:23:01.444 | INFO | acestep.core.generation.handler.service_generate_execute:_execute_service_generate_diffusion:263 - [service_generate] DiT diffusion via PyTorch (cuda)... [Python] Generating (acestep-v15-xl-sft, 50 steps)... Using precomputed LM hints Using precomputed LM hints 0%| | 0/50 [00:00