20261023_h3_FL2VA_0.4mp10s_omnixpu加速_B70_64G_冷启+热启.txt · 2026-10-23 02:52
| 阶段 | 耗时 | 步数 | s/it | staged MB | 其他指标 |
|---|---|---|---|---|---|
| 冷启 | 211.47 s | 4 | 27.23 | 25,882 | CUTE_BHLD_D128 ×3CUTE_H3_VAE_D64 ×3 |
| 热启 | 182.45 s | 4 | 27.64 | 25,882 | — |
[INFO] got prompt [INFO] VAE load device: xpu:0, offload device: cpu, dtype: torch.float32 [INFO] VAE load device: xpu:0, offload device: cpu, dtype: torch.float16 [INFO] Found quantization metadata version 1 [INFO] Using MixedPrecisionOps for text encoder [INFO] CLIP/text encoder model load device: xpu:0, offload device: cpu, current: cpu, dtype: torch.float16 [INFO] Requested to load MiniMaxH3TEModel_ [INFO] Model MiniMaxH3TEModel_ prepared for dynamic VRAM loading. 25882MB Staged. 0 patches attached. Force pre-loaded 310 weights: 1272 KB. [INFO] [OmniXPU] norm first use: LayerNorm shape=torch.Size([1620, 1152]) [INFO] [OmniXPU] AIMDO inference budget: devices=0 minimum=221.8MiB [INFO] Requested to load MiniMaxH3VideoVAE [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] Found quantization metadata version 1 [INFO] Detected mixed precision quantization [INFO] Using mixed precision operations [INFO] Native ops: asym_w4a8_int8, convrot_w4a4, int8_tensorwise , emulated ops: float8_e4m3fn, mxfp8, float8_e5m2, nvfp4 [INFO] model weight dtype torch.bfloat16, manual cast: torch.bfloat16 [INFO] model_type FLOW_AV [INFO] [OmniXPU] LoRA memory budget cached: source=1865.5MiB/416 tensors model_added=1865.5MiB model_budget=1865.5MiB clip_added=0.0MiB clip_budget=0.0MiB [INFO] [OmniXPU] LoRA model budget applied: cached=1865.5MiB/416 tensors requested=12883.3->14748.8MiB minimum=6441.6->8307.1MiB xpu_allocated=1.9MiB xpu_reserved=2.0MiB xpu_free=26995.1MiB [INFO] [OmniXPU] AIMDO inference budget: devices=0 minimum=8307.1MiB [INFO] Requested to load MiniMaxH3 [INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 208 patches attached. Force pre-loaded 210 weights: 1175 KB. [INFO] [OmniXPU] LoRA model load complete: xpu_allocated=3.1MiB xpu_reserved=4.0MiB xpu_free=26995.2MiB [WARNING] [OmniXPU] experimental capability-driven BMG D128 CUTE route enabled; set OMNI_ATTN_BACKEND=torch to disable it if this workflow shows an issue [INFO] [OmniXPU] attention CUTE_BHLD_D128 #1: heads=56 q=419 kv=419 dtype=torch.bfloat16 [INFO] [OmniXPU] attention CUTE_BHLD_D128 #2: heads=56 q=419 kv=419 dtype=torch.bfloat16 0%| | 0/4 [00:00<?, ?it/s, Model Initializing ... ][INFO] [OmniXPU] attention CUTE_BHLD_D128 #3: heads=56 q=30794 kv=30794 dtype=torch.bfloat16 100%|██████████| 4/4 [01:48<00:00, 27.23s/it] [INFO] [OmniXPU] AIMDO inference budget: devices=0 minimum=358.7MiB [INFO] Requested to load MiniMaxH3AudioVAE [INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB. [INFO] [OmniXPU] AIMDO inference budget: devices=0 minimum=762.6MiB [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] [OmniXPU] attention CUTE_H3_VAE_D64 #1: heads=32 seq=1797 dtype=torch.float16 [INFO] [OmniXPU] attention CUTE_H3_VAE_D64 #2: heads=32 seq=1797 dtype=torch.float16 [INFO] [OmniXPU] attention CUTE_H3_VAE_D64 #3: heads=32 seq=1797 dtype=torch.float16 [INFO] Prompt executed in 211.47 seconds [INFO] got prompt [INFO] Model MiniMaxH3TEModel_ prepared for dynamic VRAM loading. 25882MB Staged. 0 patches attached. Force pre-loaded 310 weights: 1272 KB. [INFO] [OmniXPU] AIMDO inference budget: devices=0 minimum=221.8MiB [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] [OmniXPU] LoRA model budget applied: cached=1865.5MiB/416 tensors requested=12883.3->14748.8MiB minimum=6441.6->8307.1MiB xpu_allocated=4.0MiB xpu_reserved=4.0MiB xpu_free=18443.0MiB [INFO] [OmniXPU] AIMDO inference budget: devices=0 minimum=8307.1MiB [INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 19995MB Staged. 208 patches attached. Force pre-loaded 210 weights: 1175 KB. [INFO] [OmniXPU] LoRA model load complete: xpu_allocated=4.0MiB xpu_reserved=4.0MiB xpu_free=18407.1MiB 100%|██████████| 4/4 [01:50<00:00, 27.64s/it] [INFO] [OmniXPU] AIMDO inference budget: devices=0 minimum=358.7MiB [INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB. [INFO] [OmniXPU] AIMDO inference budget: devices=0 minimum=762.6MiB [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] Prompt executed in 182.45 seconds