2026-01-07 09:55:53.359 | INFO     | indextts.infer_vllm_v2:__init__:105 - >> GPT weights restored from: checkpoints/IndexTTS-2-vLLM/gpt.pth
2026-01-07 09:55:53.372 | INFO     | indextts.infer_vllm_v2:__init__:116 - >> Failed to load custom CUDA kernel for BigVGAN. Falling back to torch.
2026-01-07 09:55:56.908 | INFO     | indextts.infer_vllm_v2:__init__:137 - >> semantic_codec weights restored from: checkpoints/IndexTTS-2-vLLM/semantic_codec/model.safetensors
2026-01-07 10:08:12.331 | INFO     | indextts.infer_vllm_v2:__init__:105 - >> GPT weights restored from: checkpoints/IndexTTS-2-vLLM/gpt.pth
2026-01-07 10:08:12.340 | INFO     | indextts.infer_vllm_v2:__init__:116 - >> Failed to load custom CUDA kernel for BigVGAN. Falling back to torch.
2026-01-07 10:08:15.863 | INFO     | indextts.infer_vllm_v2:__init__:137 - >> semantic_codec weights restored from: checkpoints/IndexTTS-2-vLLM/semantic_codec/model.safetensors
2026-01-07 10:08:18.419 | INFO     | indextts.infer_vllm_v2:__init__:152 - >> s2mel weights restored from: checkpoints/IndexTTS-2-vLLM/s2mel.pth
2026-01-07 10:08:18.781 | INFO     | indextts.infer_vllm_v2:__init__:163 - >> campplus_model weights restored from: checkpoints/IndexTTS-2-vLLM/campplus/campplus_cn_common.bin
2026-01-07 10:08:22.673 | INFO     | indextts.infer_vllm_v2:__init__:171 - >> bigvgan weights restored from: nvidia/bigvgan_v2_22khz_80band_256x
2026-01-07 10:13:42.833 | INFO     | indextts.infer_vllm_v2:__init__:105 - >> GPT weights restored from: checkpoints/IndexTTS-2-vLLM/gpt.pth
2026-01-07 10:13:42.842 | INFO     | indextts.infer_vllm_v2:__init__:116 - >> Failed to load custom CUDA kernel for BigVGAN. Falling back to torch.
2026-01-07 10:13:46.408 | INFO     | indextts.infer_vllm_v2:__init__:137 - >> semantic_codec weights restored from: checkpoints/IndexTTS-2-vLLM/semantic_codec/model.safetensors
2026-01-07 10:13:48.547 | INFO     | indextts.infer_vllm_v2:__init__:152 - >> s2mel weights restored from: checkpoints/IndexTTS-2-vLLM/s2mel.pth
2026-01-07 10:13:48.892 | INFO     | indextts.infer_vllm_v2:__init__:163 - >> campplus_model weights restored from: checkpoints/IndexTTS-2-vLLM/campplus/campplus_cn_common.bin
2026-01-07 10:13:52.163 | INFO     | indextts.infer_vllm_v2:__init__:171 - >> bigvgan weights restored from: nvidia/bigvgan_v2_22khz_80band_256x
2026-01-07 11:35:01.871 | INFO     | indextts.infer_vllm_v2:__init__:105 - >> GPT weights restored from: checkpoints/IndexTTS-2-vLLM/gpt.pth
2026-01-07 11:35:01.882 | INFO     | indextts.infer_vllm_v2:__init__:116 - >> Failed to load custom CUDA kernel for BigVGAN. Falling back to torch.
2026-01-07 11:35:05.240 | INFO     | indextts.infer_vllm_v2:__init__:137 - >> semantic_codec weights restored from: checkpoints/IndexTTS-2-vLLM/semantic_codec/model.safetensors
2026-01-07 11:35:07.533 | INFO     | indextts.infer_vllm_v2:__init__:152 - >> s2mel weights restored from: checkpoints/IndexTTS-2-vLLM/s2mel.pth
2026-01-07 11:35:07.892 | INFO     | indextts.infer_vllm_v2:__init__:163 - >> campplus_model weights restored from: checkpoints/IndexTTS-2-vLLM/campplus/campplus_cn_common.bin
2026-01-07 11:35:11.961 | INFO     | indextts.infer_vllm_v2:__init__:171 - >> bigvgan weights restored from: nvidia/bigvgan_v2_22khz_80band_256x
2026-01-07 11:53:45.877 | INFO     | indextts.infer_vllm_v2:__init__:105 - >> GPT weights restored from: checkpoints/IndexTTS-2-vLLM/gpt.pth
2026-01-07 11:53:45.889 | INFO     | indextts.infer_vllm_v2:__init__:116 - >> Failed to load custom CUDA kernel for BigVGAN. Falling back to torch.
2026-01-07 11:53:49.144 | INFO     | indextts.infer_vllm_v2:__init__:137 - >> semantic_codec weights restored from: checkpoints/IndexTTS-2-vLLM/semantic_codec/model.safetensors
2026-01-07 11:53:51.373 | INFO     | indextts.infer_vllm_v2:__init__:152 - >> s2mel weights restored from: checkpoints/IndexTTS-2-vLLM/s2mel.pth
2026-01-07 11:53:51.762 | INFO     | indextts.infer_vllm_v2:__init__:163 - >> campplus_model weights restored from: checkpoints/IndexTTS-2-vLLM/campplus/campplus_cn_common.bin
2026-01-07 11:53:55.438 | INFO     | indextts.infer_vllm_v2:__init__:171 - >> bigvgan weights restored from: nvidia/bigvgan_v2_22khz_80band_256x
2026-01-07 11:54:32.456 | INFO     | indextts.infer_vllm_v2:__init__:176 - >> TextNormalizer loaded
2026-01-07 11:54:32.484 | INFO     | indextts.infer_vllm_v2:__init__:178 - >> bpe model loaded from: checkpoints/IndexTTS-2-vLLM/bpe.model
2026-01-07 12:26:15.303 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 12:26:33.498 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 12:26:34.437 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [9e538135d39f4ee1b990f7940806b044] [prefill time: 0.9366]
2026-01-07 12:26:46.639 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [9e538135d39f4ee1b990f7940806b044] [decode time: 12.2017] [decode len: 1761]
2026-01-07 12:26:47.435 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 12:26:47.855 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 12:26:47.874 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [1c403009d06c4c3c9811620525f339aa] [prefill time: 0.0177]
2026-01-07 12:26:59.764 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [1c403009d06c4c3c9811620525f339aa] [decode time: 11.8893] [decode len: 1761]
2026-01-07 12:27:00.146 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 12:27:00.777 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 12:27:00.801 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [d8300dd0b6b749b08352ea5468732605] [prefill time: 0.0174]
2026-01-07 12:27:12.761 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [d8300dd0b6b749b08352ea5468732605] [decode time: 11.9596] [decode len: 1761]
2026-01-07 12:27:13.116 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 12:27:14.408 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 12:27:14.432 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [f0d11d0427cd4ac1a83f89d987d137e3] [prefill time: 0.0176]
2026-01-07 12:27:15.221 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [f0d11d0427cd4ac1a83f89d987d137e3] [decode time: 0.7887] [decode len: 135]
2026-01-07 15:39:20.324 | INFO     | indextts.infer_vllm_v2:__init__:105 - >> GPT weights restored from: checkpoints/IndexTTS-2-vLLM/gpt.pth
2026-01-07 15:39:20.332 | INFO     | indextts.infer_vllm_v2:__init__:116 - >> Failed to load custom CUDA kernel for BigVGAN. Falling back to torch.
2026-01-07 15:39:23.893 | INFO     | indextts.infer_vllm_v2:__init__:137 - >> semantic_codec weights restored from: checkpoints/IndexTTS-2-vLLM/semantic_codec/model.safetensors
2026-01-07 15:39:26.238 | INFO     | indextts.infer_vllm_v2:__init__:152 - >> s2mel weights restored from: checkpoints/IndexTTS-2-vLLM/s2mel.pth
2026-01-07 15:39:26.591 | INFO     | indextts.infer_vllm_v2:__init__:163 - >> campplus_model weights restored from: checkpoints/IndexTTS-2-vLLM/campplus/campplus_cn_common.bin
2026-01-07 15:39:30.100 | INFO     | indextts.infer_vllm_v2:__init__:171 - >> bigvgan weights restored from: nvidia/bigvgan_v2_22khz_80band_256x
2026-01-07 15:39:32.076 | INFO     | indextts.infer_vllm_v2:__init__:176 - >> TextNormalizer loaded
2026-01-07 15:39:32.108 | INFO     | indextts.infer_vllm_v2:__init__:178 - >> bpe model loaded from: checkpoints/IndexTTS-2-vLLM/bpe.model
2026-01-07 15:40:39.322 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 15:40:45.193 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 15:40:46.224 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [2c7b38cf2739452c9ebdab36aa5c4db0] [prefill time: 1.0290]
2026-01-07 15:40:48.352 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [2c7b38cf2739452c9ebdab36aa5c4db0] [decode time: 2.1280] [decode len: 336]
2026-01-07 15:40:48.576 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 15:40:48.999 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 15:40:49.018 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [00454b8a816048a998374d8b6dd2df81] [prefill time: 0.0177]
2026-01-07 15:40:53.017 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [00454b8a816048a998374d8b6dd2df81] [decode time: 3.9981] [decode len: 616]
2026-01-07 15:40:53.215 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 15:40:53.674 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 15:40:53.696 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [eced97d4f81d4023a9406bf25f2af326] [prefill time: 0.0200]
2026-01-07 15:40:56.550 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [eced97d4f81d4023a9406bf25f2af326] [decode time: 2.8534] [decode len: 439]
2026-01-07 15:40:56.712 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 15:40:57.999 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 15:40:58.018 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [36cdd0e5c0fa4b31b80e1733fc2aa77b] [prefill time: 0.0171]
2026-01-07 15:40:58.186 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [36cdd0e5c0fa4b31b80e1733fc2aa77b] [decode time: 0.1675] [decode len: 29]
2026-01-07 17:16:04.466 | INFO     | indextts.infer_vllm_v2:__init__:105 - >> GPT weights restored from: checkpoints/IndexTTS-2-vLLM/gpt.pth
2026-01-07 17:16:04.473 | INFO     | indextts.infer_vllm_v2:__init__:116 - >> Failed to load custom CUDA kernel for BigVGAN. Falling back to torch.
2026-01-07 17:16:07.821 | INFO     | indextts.infer_vllm_v2:__init__:137 - >> semantic_codec weights restored from: checkpoints/IndexTTS-2-vLLM/semantic_codec/model.safetensors
2026-01-07 17:16:09.645 | INFO     | indextts.infer_vllm_v2:__init__:152 - >> s2mel weights restored from: checkpoints/IndexTTS-2-vLLM/s2mel.pth
2026-01-07 17:16:09.974 | INFO     | indextts.infer_vllm_v2:__init__:163 - >> campplus_model weights restored from: checkpoints/IndexTTS-2-vLLM/campplus/campplus_cn_common.bin
2026-01-07 17:16:13.027 | INFO     | indextts.infer_vllm_v2:__init__:171 - >> bigvgan weights restored from: nvidia/bigvgan_v2_22khz_80band_256x
2026-01-07 17:16:15.973 | INFO     | indextts.infer_vllm_v2:__init__:176 - >> TextNormalizer loaded
2026-01-07 17:16:16.002 | INFO     | indextts.infer_vllm_v2:__init__:178 - >> bpe model loaded from: checkpoints/IndexTTS-2-vLLM/bpe.model
2026-01-07 17:17:34.294 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 17:17:40.179 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 17:17:41.249 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [22f7281eaa15416488635de965f49159] [prefill time: 1.0658]
2026-01-07 17:17:52.958 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [22f7281eaa15416488635de965f49159] [decode time: 11.7088] [decode len: 1761]
2026-01-07 17:18:06.446 | INFO     | indextts.infer_vllm_v2:infer:458 - >> gpt_gen_time: 12.86 seconds
2026-01-07 17:18:06.446 | INFO     | indextts.infer_vllm_v2:infer:459 - >> gpt_forward_time: 0.18 seconds
2026-01-07 17:18:06.447 | INFO     | indextts.infer_vllm_v2:infer:460 - >> s2mel_time: 11.15 seconds
2026-01-07 17:18:06.448 | INFO     | indextts.infer_vllm_v2:infer:461 - >> bigvgan_time: 2.13 seconds
2026-01-07 17:18:06.448 | INFO     | indextts.infer_vllm_v2:infer:462 - >> Total inference time: 32.13 seconds
2026-01-07 17:18:06.449 | INFO     | indextts.infer_vllm_v2:infer:463 - >> Generated audio length: 35.12 seconds
2026-01-07 17:18:06.449 | INFO     | indextts.infer_vllm_v2:infer:464 - >> RTF: 0.9149
2026-01-07 17:18:06.465 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 17:18:06.856 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 17:18:06.875 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [4bee66e70af64239a75fa180dbe00801] [prefill time: 0.0163]
2026-01-07 17:18:18.407 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [4bee66e70af64239a75fa180dbe00801] [decode time: 11.5326] [decode len: 1761]
2026-01-07 17:18:29.940 | INFO     | indextts.infer_vllm_v2:infer:458 - >> gpt_gen_time: 11.62 seconds
2026-01-07 17:18:29.941 | INFO     | indextts.infer_vllm_v2:infer:459 - >> gpt_forward_time: 0.18 seconds
2026-01-07 17:18:29.942 | INFO     | indextts.infer_vllm_v2:infer:460 - >> s2mel_time: 9.59 seconds
2026-01-07 17:18:29.942 | INFO     | indextts.infer_vllm_v2:infer:461 - >> bigvgan_time: 0.08 seconds
2026-01-07 17:18:29.942 | INFO     | indextts.infer_vllm_v2:infer:462 - >> Total inference time: 23.45 seconds
2026-01-07 17:18:29.942 | INFO     | indextts.infer_vllm_v2:infer:463 - >> Generated audio length: 35.12 seconds
2026-01-07 17:18:29.942 | INFO     | indextts.infer_vllm_v2:infer:464 - >> RTF: 0.6677
2026-01-07 17:18:29.960 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 17:18:30.355 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 17:18:30.377 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [991824af0b334ef797bfd3a59d5cced6] [prefill time: 0.0148]
2026-01-07 17:18:35.654 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [991824af0b334ef797bfd3a59d5cced6] [decode time: 5.2770] [decode len: 876]
2026-01-07 17:18:42.032 | INFO     | indextts.infer_vllm_v2:infer:458 - >> gpt_gen_time: 5.34 seconds
2026-01-07 17:18:42.032 | INFO     | indextts.infer_vllm_v2:infer:459 - >> gpt_forward_time: 0.03 seconds
2026-01-07 17:18:42.034 | INFO     | indextts.infer_vllm_v2:infer:460 - >> s2mel_time: 5.48 seconds
2026-01-07 17:18:42.034 | INFO     | indextts.infer_vllm_v2:infer:461 - >> bigvgan_time: 0.84 seconds
2026-01-07 17:18:42.034 | INFO     | indextts.infer_vllm_v2:infer:462 - >> Total inference time: 12.04 seconds
2026-01-07 17:18:42.034 | INFO     | indextts.infer_vllm_v2:infer:463 - >> Generated audio length: 17.45 seconds
2026-01-07 17:18:42.035 | INFO     | indextts.infer_vllm_v2:infer:464 - >> RTF: 0.6902
2026-01-07 17:18:42.049 | INFO     | indextts.infer_vllm_v2:infer:243 - >> start inference...
2026-01-07 17:18:43.081 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:222 - Use the specified emotion vector
2026-01-07 17:18:43.099 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:248 - [8382f6c343604dcab57888aa85574f6c] [prefill time: 0.0162]
2026-01-07 17:18:54.859 | INFO     | indextts.gpt.model_vllm_v2:inference_speech:251 - [8382f6c343604dcab57888aa85574f6c] [decode time: 11.7604] [decode len: 1761]
2026-01-07 17:19:06.392 | INFO     | indextts.infer_vllm_v2:infer:458 - >> gpt_gen_time: 11.84 seconds
2026-01-07 17:19:06.393 | INFO     | indextts.infer_vllm_v2:infer:459 - >> gpt_forward_time: 0.04 seconds
2026-01-07 17:19:06.393 | INFO     | indextts.infer_vllm_v2:infer:460 - >> s2mel_time: 9.74 seconds
2026-01-07 17:19:06.394 | INFO     | indextts.infer_vllm_v2:infer:461 - >> bigvgan_time: 0.08 seconds
2026-01-07 17:19:06.395 | INFO     | indextts.infer_vllm_v2:infer:462 - >> Total inference time: 24.32 seconds
2026-01-07 17:19:06.395 | INFO     | indextts.infer_vllm_v2:infer:463 - >> Generated audio length: 35.12 seconds
2026-01-07 17:19:06.395 | INFO     | indextts.infer_vllm_v2:infer:464 - >> RTF: 0.6925