support HF_TOKEN

Author: ngxson

convert_hf_to_gguf.py

@@ -5498,7 +5498,7 @@ def parse_args() -> argparse.Namespace:
     )
     parser.add_argument(
         "--remote", action="store_true",
-        help="(Experimental) Read safetensors file remotely without downloading to disk. Config and tokenizer files will still be downloaded. To use this feature, you need to specify Hugging Face model repo name instead of a local directory. For example: 'HuggingFaceTB/SmolLM2-1.7B-Instruct'",
+        help="(Experimental) Read safetensors file remotely without downloading to disk. Config and tokenizer files will still be downloaded. To use this feature, you need to specify Hugging Face model repo name instead of a local directory. For example: 'HuggingFaceTB/SmolLM2-1.7B-Instruct'. Note: To access gated repo, set HF_TOKEN environment variable to your Hugging Face token.",
     )
 
     args = parser.parse_args()

gguf-py/gguf/utility.py

@@ -3,6 +3,7 @@
 from dataclasses import dataclass
 from typing import Literal
 
+import os
 import json
 
 
@@ -94,7 +95,7 @@ class SafetensorRemote:
     Example (one model has single safetensor file, the other has multiple):
         for model_id in ["ngxson/TEST-Tiny-Llama4", "Qwen/Qwen2.5-7B-Instruct"]:
             tensors = SafetensorRemote.get_list_tensors_hf_model(model_id)
-            print(json.dumps(tensors, indent=2))
+            print(tensors)
 
     Example reading tensor data:
         tensors = SafetensorRemote.get_list_tensors_hf_model(model_id)
@@ -223,8 +224,10 @@ def get_data_by_range(cls, url: str, start: int, size: int = -1) -> bytes:
             raise ValueError(f"Invalid URL: {url}")
 
         headers = {}
+        if os.environ.get("HF_TOKEN"):
+            headers["Authorization"] = f"Bearer {os.environ['HF_TOKEN']}"
         if size > -1:
-            headers = {"Range": f"bytes={start}-{start + size}"}
+            headers["Range"] = f"bytes={start}-{start + size}"
         response = requests.get(url, allow_redirects=True, headers=headers)
         response.raise_for_status()
 
@@ -246,6 +249,8 @@ def check_file_exist(cls, url: str) -> bool:
 
         try:
             headers = {"Range": "bytes=0-0"}
+            if os.environ.get("HF_TOKEN"):
+                headers["Authorization"] = f"Bearer {os.environ['HF_TOKEN']}"
             response = requests.head(url, allow_redirects=True, headers=headers)
             # Success (2xx) or redirect (3xx)
             return 200 <= response.status_code < 400