Add vLLM-Omni library info to Hugging Face Hub requests (#5381)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
authored and GitHub committed 2026-09-08 10:08:38 -06:00
1 parent 7a80b79698
commit fdaf4e31bd
57 files changed
+259 -220

No files matched your search

@@ -1,5 +1,5 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project
"""External model loading, audio I/O, and emotion conditioning for IndexTTS2."""
from __future__ import annotations
@@ -171,9 +171,10 @@ def load_semantic_codec(
codec.load_state_dict(state, strict=False)
else:
import safetensors.torch
from huggingface_hub import hf_hub_download
ckpt_path = hf_hub_download("amphion/MaskGCT", filename="semantic_codec/model.safetensors")
from vllm_omni.transformers_utils.repo_utils import hf_api
ckpt_path = hf_api().hf_hub_download("amphion/MaskGCT", filename="semantic_codec/model.safetensors")
safetensors.torch.load_model(codec, ckpt_path)
_freeze(codec.to(device=device, dtype=torch.float32))
_semantic_codec_cache[cache_key] = codec
@@ -198,9 +199,11 @@ def load_campplus(model_path: str, device: torch.device):
if ckpt_path is not None:
break
if ckpt_path is None:
from huggingface_hub import hf_hub_download
if os.path.isdir(model_path):
raise FileNotFoundError(f"IndexTTS CAMPPlus checkpoint is missing from local bundle {model_path!r}")
from vllm_omni.transformers_utils.repo_utils import hf_api
ckpt_path = hf_hub_download("funasr/campplus", filename="campplus_cn_common.bin")
ckpt_path = hf_api().hf_hub_download("funasr/campplus", filename="campplus_cn_common.bin")
state = torch.load(ckpt_path, map_location="cpu", weights_only=True)
campplus.load_state_dict(state, strict=False)
_freeze(campplus.to(device=device, dtype=torch.float32))
@@ -1,3 +1,6 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project
# Copyright (c) 2024 NVIDIA CORPORATION.
# Licensed under the MIT license.
@@ -9,13 +12,14 @@ import os
import torch
import torch.nn as nn
from huggingface_hub import PyTorchModelHubMixin, hf_hub_download
from huggingface_hub import PyTorchModelHubMixin
from torch.nn import Conv1d, ConvTranspose1d
from torch.nn.utils import remove_weight_norm, weight_norm
from vllm_omni.model_executor.models.common.alias_free_activation import AliasFreeActivation1d
from vllm_omni.model_executor.models.common.snake_activation import Snake, SnakeBeta
from vllm_omni.model_executor.models.indextts2.s2mel.modules.commons import AttrDict
from vllm_omni.transformers_utils.repo_utils import hf_api
# ---------------------------------------------------------------------------
# Helpers (inlined from env.py / utils.py)
@@ -328,7 +332,7 @@ class BigVGAN(
def _resolve(filename):
if os.path.isdir(model_id):
return os.path.join(model_id, filename)
return hf_hub_download(
return hf_api().hf_hub_download(
repo_id=model_id,
filename=filename,
**{
@@ -1,5 +1,5 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project
"""IndexTTS 2.5 multilingual tiktoken vocabulary.
The vocabulary layout mirrors the official IndexTTS 2.5 tokenizer: 58,836
@@ -234,9 +234,9 @@ def resolve_indextts25_tokenizer_file(
return str(model_path)
try:
from huggingface_hub import hf_hub_download
from vllm_omni.transformers_utils.repo_utils import hf_api
return hf_hub_download(repo_id=model_dir, filename=tokenizer_file)
return hf_api().hf_hub_download(repo_id=model_dir, filename=tokenizer_file)
except Exception as exc:
raise FileNotFoundError(
f"Could not resolve IndexTTS 2.5 tokenizer {tokenizer_file!r} from {model_dir!r}"