A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 8, 17:45 UTC
[Docs] Improve documentation (#33799)
Co-authored-by: Soren Dreano <soren@numind.ai> Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
SorenDreano committed
Feb 6, 2026 at 13:57 UTC
6e7b1c4b591a7d735fc93792e53cb5592cfab4f2
4 files changed
+12
-12
docs/models/supported_models.md
+8
-8
@@ -224,13 +224,13 @@ If you prefer, you can use the Hugging Face CLI to [download a model](https://hu
224
225
```bash
226
# Download a model
227
-huggingface-cli download HuggingFaceH4/zephyr-7b-beta
227
+hf download HuggingFaceH4/zephyr-7b-beta
228
229
# Specify a custom cache directory
230
-huggingface-cli download HuggingFaceH4/zephyr-7b-beta --cache-dir ./path/to/cache
230
+hf download HuggingFaceH4/zephyr-7b-beta --cache-dir ./path/to/cache
231
232
# Download a specific file from a model repo
233
-huggingface-cli download HuggingFaceH4/zephyr-7b-beta eval_results.json
233
+hf download HuggingFaceH4/zephyr-7b-beta eval_results.json
234
```
235
236
#### List the downloaded models
@@ -239,13 +239,13 @@ Use the Hugging Face CLI to [manage models](https://huggingface.co/docs/huggingf
239
240
```bash
241
# List cached models
242
-huggingface-cli scan-cache
242
+hf scan-cache
243
244
# Show detailed (verbose) output
245
-huggingface-cli scan-cache -v
245
+hf scan-cache -v
246
247
# Specify a custom cache directory
248
-huggingface-cli scan-cache --dir ~/.cache/huggingface/hub
248
+hf scan-cache --dir ~/.cache/huggingface/hub
249
```
250
251
#### Delete a cached model
@@ -260,7 +260,7 @@ Use the Hugging Face CLI to interactively [delete downloaded model](https://hugg
260
# Please run `pip install huggingface_hub[cli]` to install them.
261
262
# Launch the interactive TUI to select models to delete
263
-$ huggingface-cli delete-cache
263
+$ hf delete-cache
264
? Select revisions to delete: 1 revisions selected counting for 438.9M.
265
○ None of the following (if selected, nothing will be deleted).
266
Model BAAI/bge-base-en-v1.5 (438.9M, used 1 week ago)
@@ -297,7 +297,7 @@ export https_proxy=http://your.proxy.server:port
297
- Set the proxy for just the current command:
298
299
```shell
300
-https_proxy=http://your.proxy.server:port huggingface-cli download <model_name>
300
+https_proxy=http://your.proxy.server:port hf download <model_name>
301
302
# or use vllm cmd directly
303
https_proxy=http://your.proxy.server:port vllm serve <model_name>
examples/offline_inference/openai_batch/README.md
+1
-1
@@ -20,7 +20,7 @@ We currently support `/v1/chat/completions`, `/v1/embeddings`, and `/v1/score` e
20
21
* The examples in this document use `meta-llama/Meta-Llama-3-8B-Instruct`.
22
* Create a [user access token](https://huggingface.co/docs/hub/en/security-tokens)
23
- * Install the token on your machine (Run `huggingface-cli login`).
23
+ * Install the token on your machine (Run `hf auth login`).
24
* Get access to the gated model by [visiting the model card](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) and agreeing to the terms and conditions.
25
26
## Example 1: Running with a local file
vllm/config/model.py
+2
-2
@@ -247,8 +247,8 @@ class ModelConfig:
247
- "mistral" will load the config in mistral format."""
248
hf_token: bool | str | None = None
249
"""The token to use as HTTP bearer authorization for remote files . If
250
- `True`, will use the token generated when running `huggingface-cli login`
251
- (stored in `~/.huggingface`)."""
250
+ `True`, will use the token generated when running `hf auth login`
251
+ (stored in `~/.cache/huggingface/token`)."""
252
hf_overrides: HfOverrides = field(default_factory=dict)
253
"""If a dictionary, contains arguments to be forwarded to the Hugging Face
254
config. If a callable, it is called to update the HuggingFace config."""
vllm/entrypoints/llm.py
+1
-1
@@ -170,7 +170,7 @@ class LLM:
170
[ParallelConfig][vllm.config.ParallelConfig].
171
hf_token: The token to use as HTTP bearer authorization for remote files
172
. If `True`, will use the token generated when running
173
- `huggingface-cli login` (stored in `~/.huggingface`).
173
+ `hf auth login` (stored in `~/.cache/huggingface/token`).
174
hf_overrides: If a dictionary, contains arguments to be forwarded to the
175
HuggingFace config. If a callable, it is called to update the
176
HuggingFace config.