๐Ÿ“ข [v0.23.0]: LLMs with tools, seamless downloads, and much more!

#6
by Wauplin HF staff - opened
Owner
โ€ข
edited May 2

Note: pre-release 0.23.0.rc0 is available on PyPI. Official release will occur in the coming days.
EDIT: 0.23.0 is out!

๐Ÿ“ Seamless download to local dir

The 0.23.0 release comes with a big revamp of the download process, especially when it comes to downloading to a local directory. Previously the process was still involving the cache directory and symlinks which led to misconceptions and a suboptimal user experience. The new workflow involves a .cache/huggingface/ folder, similar to the .git/ one, that keeps track of the progress of a download. The main features are:

  • no symlinks
  • no local copy
  • don't re-download when not necessary
  • same behavior on both Unix and Windows
  • unrelated to cache-system

Example to download q4 GGUF file for microsoft/Phi-3-mini-4k-instruct-gguf:

# Download q4 GGUF file from 
huggingface-cli download microsoft/Phi-3-mini-4k-instruct-gguf Phi-3-mini-4k-instruct-q4.gguf --local-dir=data/phi3

With this addition, interrupted downloads are now resumable! This applies both for downloads in local and cache directories which should greatly improve UX for users with slow/unreliable connections. In this regard, the resume_download parameter is now deprecated (not relevant anymore).

  • Revamp download to local dir process by @Wauplin in #2223
  • Rename .huggingface/ folder to .cache/huggingface/ by @Wauplin in #2262

๐Ÿ’ก Grammar and Tools in InferenceClient

It is now possible to provide a list of tools when chatting with a model using the InferenceClient! This major improvement has been made possible thanks to TGI that handle them natively.

>>> from huggingface_hub import InferenceClient

# Ask for weather in the next days using tools
>>> client = InferenceClient("meta-llama/Meta-Llama-3-70B-Instruct")
>>> messages = [
...     {"role": "system", "content": "Don't make assumptions about what values to plug into functions. Ask for clarification if a user request is ambiguous."},
...     {"role": "user", "content": "What's the weather like the next 3 days in San Francisco, CA?"},
... ]
>>> tools = [
...     {
...         "type": "function",
...         "function": {
...             "name": "get_current_weather",
...             "description": "Get the current weather",
...             "parameters": {
...                 "type": "object",
...                 "properties": {
...                     "location": {
...                         "type": "string",
...                         "description": "The city and state, e.g. San Francisco, CA",
...                     },
...                     "format": {
...                         "type": "string",
...                         "enum": ["celsius", "fahrenheit"],
...                         "description": "The temperature unit to use. Infer this from the users location.",
...                     },
...                 },
...                 "required": ["location", "format"],
...             },
...         },
...     },
...     ...
... ]
>>> response = client.chat_completion(
...     model="meta-llama/Meta-Llama-3-70B-Instruct",
...     messages=messages,
...     tools=tools,
...     tool_choice="auto",
...     max_tokens=500,
... )
>>> response.choices[0].message.tool_calls[0].function
ChatCompletionOutputFunctionDefinition(
    arguments={
        'location': 'San Francisco, CA',
        'format': 'fahrenheit',
        'num_days': 3
    },
    name='get_n_day_weather_forecast',
    description=None
)

It is also possible to provide grammar rules to the text_generation task. This ensures that the output follows a precise JSON Schema specification or matches a regular expression. For more details about it, check out the Guidance guide from Text-Generation-Inference docs.

  • Add support for Grammar/Tools + TGI-based specs in InferenceClient by @Wauplin in #2237

โš™๏ธ Other

Mention more chat-completion task instead of conversation in documentation.

  • Add chat_completion and remove conversational from Inference guide by @Wauplin in #2215

chat-completion relies on server-side rendering in all cases, including when model is transformers-backed. Previously it was only the case for TGI-backed models and templates were rendered client-side otherwise.

  • Render chat-template server-side for transformers-backed models by @Wauplin in #2258

Improved logic to determine whether a model is served via TGI or transformers.

๐ŸŒ ๐Ÿ“š Korean community is on fire!

The PseudoLab team is a non-profit dedicated to make AI more accessible in the Korean-speaking community. In the past few weeks, their team of contributors managed to translated (almost) entirely the huggingface_hub documentation. Huge shout-out to the coordination on this task! Documentation can be accessed here.

  • ๐ŸŒ [i18n-KO] Translated guides/webhooks_server.md to Korean by @nuatmochoi in #2145
  • ๐ŸŒ [i18n-KO] Translated reference/login.md to Korean by @SeungAhSon in #2151
  • ๐ŸŒ [i18n-KO] Translated package_reference/tensorboard.md to Korean by @fabxoe in #2173
  • ๐ŸŒ [i18n-KO] Translated package_reference/inference_client.md to Korean by @cjfghk5697 in #2178
  • ๐ŸŒ [i18n-KO] Translated reference/inference_endpoints.md to Korean by @harheem in #2180
  • ๐ŸŒ [i18n-KO] Translated package_reference/file_download.md to Korean by @seoyoung-3060 in #2184
  • ๐ŸŒ [i18n-KO] Translated package_reference/cache.md to Korean by @nuatmochoi in #2191
  • ๐ŸŒ [i18n-KO] Translated package_reference/collections.md to Korean by @boyunJang in #2214
  • ๐ŸŒ [i18n-KO] Translated package_reference/inference_types.md to Korean by @fabxoe in #2171
  • ๐ŸŒ [i18n-KO] Translated guides/upload.md to Korean by @junejae in #2139
  • ๐ŸŒ [i18n-KO] Translated reference/repository.md to Korean by @junejae in #2189
  • ๐ŸŒ [i18n-KO] Translated package_reference/space_runtime.md to Korean by @boyunJang in #2213
  • ๐ŸŒ [i18n-KO] Translated guides/repository.md to Korean by @cjfghk5697 in #2124
  • ๐ŸŒ [i18n-KO] Translated guides/model_cards.md to Korean" by @SeungAhSon in #2128
  • ๐ŸŒ [i18n-KO] Translated guides/community.md to Korean by @seoulsky-field in #2126
  • ๐ŸŒ [i18n-KO] Translated guides/cli.md to Korean by @harheem in #2131
  • ๐ŸŒ [i18n-KO] Translated guides/search.md to Korean by @seoyoung-3060 in #2134
  • ๐ŸŒ [i18n-KO] Translated guides/inference.md to Korean by @boyunJang in #2130
  • ๐ŸŒ [i18n-KO] Translated guides/manage-spaces.md to Korean by @boyunJang in #2220
  • ๐ŸŒ [i18n-KO] Translating guides/hf_file_system.md to Korean by @heuristicwave in #2146
  • ๐ŸŒ [i18n-KO] Translated package_reference/hf_api.md to Korean by @fabxoe in #2165
  • ๐ŸŒ [i18n-KO] Translated package_reference/mixins.md to Korean by @fabxoe in #2166
  • ๐ŸŒ [i18n-KO] Translated guides/inference_endpoints.md to Korean by @usr-bin-ksh in #2164
  • ๐ŸŒ [i18n-KO] Translated package_reference/utilities.md to Korean by @cjfghk5697 in #2196
  • fix ko docs by @Wauplin (direct commit on main)
  • ๐ŸŒ [i18n-KO] Translated package_reference/serialization.md to Korean by @seoyoung-3060 in #2233
  • ๐ŸŒ [i18n-KO] Translated package_reference/hf_file_system.md to Korean by @SeungAhSon in #2174

๐Ÿ› ๏ธ Misc improvements

User API

@bilgehanertan added support for 2 new routes:

  • get_user_overview to retrieve high-level information about a user: username, avatar, number of models/datasets/Spaces, number of likes and upvotes, number of interactions in discussion, etc.

CLI tag

@bilgehanertan added a new command to the CLI to handle tags. It is now possible to:

  • tag a repo
>>> huggingface-cli tag Wauplin/my-cool-model v1.0
You are about to create tag v1.0 on model Wauplin/my-cool-model
Tag v1.0 created on Wauplin/my-cool-model
  • retrieve the list of tags for a repo
>>> huggingface-cli tag Wauplin/gradio-space-ci -l --repo-type space
Tags for space Wauplin/gradio-space-ci:
0.2.2
0.2.1
0.2.0
0.1.2
0.0.2
0.0.1
  • delete a tag on a repo
>>> huggingface-cli tag -d Wauplin/my-cool-model v1.0
You are about to delete tag v1.0 on model Wauplin/my-cool-model
Proceed? [Y/n] y
Tag v1.0 deleted on Wauplin/my-cool-model

For more details, check out the CLI guide.

๐Ÿงฉ ModelHubMixin

This ModelHubMixin got a set of nice improvement to generate model cards and handle custom data types in the config.json file. More info in the integration guide.

  • ModelHubMixin: more metadata + arbitrary config types + proper guide by @Wauplin in #2230
  • Fix ModelHubMixin when class is a dataclass by @Wauplin in #2159
  • Do not document private attributes of ModelHubMixin by @Wauplin in #2216
  • Add support for pipeline_tag in ModelHubMixin by @Wauplin in #2228

โš™๏ธ Other

In a shared environment, it is now possible to set a custom path HF_TOKEN_PATH as environment variable so that each user of the cluster has their own access token.

  • Support HF_TOKEN_PATH as environment variable by @Wauplin in #2185

Thanks to @Y4suyuki and @lappemic, most custom errors defined in huggingface_hub are now aggregated in the same module. This makes it very easy to import them from from huggingface_hub.errors import ....

  • Define errors in errors.py by @Y4suyuki in #2170
  • Define errors in errors file by @lappemic in #2202

Fixed HFSummaryWriter (class to seamlessly log tensorboard events to the Hub) to work with either tensorboardX or torch.utils implementation, depending on the user setup.

  • Import SummaryWriter from either tensorboardX or torch.utils by @Wauplin in #2205

Speed to list files using HfFileSystem has been drastically improved, thanks to @awgr . The values returned from the cache are not deep-copied anymore, which was unfortunately the part taking the most time in the process. If users want to modify values returned by HfFileSystem, they would need to copy them before-hand. This is expected to be a very limited drawback.

Progress bars in huggingface_hub got some flexibility!
It is now possible to provide a name to a tqdm bar (similar to logging.getLogger) and to enable/disable only some progress bars. More details in this guide.

>>> from huggingface_hub.utils import tqdm, disable_progress_bars
>>> disable_progress_bars("peft.foo")

# No progress bars for `peft.boo.bar`
>>> for _ in tqdm(range(5), name="peft.foo.bar"):
...     pass

# But for `peft` yes
>>> for _ in tqdm(range(5), name="peft"):
...     pass
100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 5/5 [00:00<00:00, 117817.53it/s]
  • Implement hierarchical progress bar control in huggingface_hub by @lappemic in #2217

๐Ÿ’” Breaking changes

--local-dir-use-symlink and --resume-download

As part of the download process revamp, some breaking changes have been introduced. However we believe that the benefits outweigh the change cost. Breaking changes include:

  • a .cache/huggingface/ folder is not present at the root of the local dir. It only contains file locks, metadata and partially downloaded files. If you need to, you can safely delete this folder without corrupting the data inside the root folder. However, you should expect a longer recovery time if you try to re-run your download command.
  • --local-dir-use-symlink is not in used anymore and will be ignored. It is not possible anymore to symlinks your local dir with the cache directory. Thanks to the .cache/huggingface/ folder, it shouldn't be needed anyway.
  • --resume-download has been deprecated and will be ignored. Resuming failed downloads is now activated by default all the time. If you need to force a new download, use --force-download.

Inference Types

As part of #2237 (Grammar and Tools support), we've updated the return value from InferenceClient.chat_completion and InferenceClient.text_generation to match exactly TGI output. The attributes of the returned objects did not change but the classes definition themselves yes. Expect errors if you've previously had from huggingface_hub import TextGenerationOutput in your code. This is however not the common usage since those objects are already instantiated by huggingface_hub directly.

Expected breaking changes

Some other breaking changes were expected (and announced since 0.19.x):

  • list_files_info is definitively removed in favor of get_paths_info and list_repo_tree
  • WebhookServer.run is definitively removed in favor of WebhookServer.launch
  • api_endpoint in ModelHubMixin push_to_hub's method is definitively removed in favor of the HF_ENDPOINT environment variable

Check #2156 for more details.

Small fixes and maintenance

โš™๏ธ CI optimization

โš™๏ธ fixes

  • Fix HF_ENDPOINT not handled correctly by @Wauplin in #2155
  • Fix proxy if dynamic endpoint by @Wauplin (direct commit on main)
  • Update the note message when logging in to make it easier to understand and clearer by @lh0x00 in #2163
  • Fix URL when uploading to proxy by @Wauplin in #2167
  • Fix SafeTensorsInfo initialization by @Wauplin in #2190
  • Doc cli download timeout by @zioalex in #2198
  • Fix Typos in CONTRIBUTION.md and Formatting in README.md by @lappemic in #2201
  • change default model card by @Wauplin (direct commit on main)
  • Add returns documentation for save_pretrained by @alexander-soare in #2226
  • Update cli.md by @QuinnPiers in #2242
  • add warning tip that list_deployed_models only searches over cache by @MoritzLaurer in #2241
  • Respect default timeouts in hf_file_system by @Wauplin in #2253
  • Update harmonized token param desc and type def by @lappemic in #2252
  • Better document download attribute by @Wauplin in #2250
  • Correctly check inference endpoint is ready by @Wauplin in #2229
  • Add support for updatedRefs in WebhookPayload by @Wauplin in #2169

โš™๏ธ internal

Significant community contributions

The following contributors have made significant changes to the library over the last release:

  • @lappemic
    • Fix Typos in CONTRIBUTION.md and Formatting in README.md (#2201)
    • Define errors in errors file (#2202)
    • [wip] Implement hierarchical progress bar control in huggingface_hub (#2217)
    • Update harmonized token param desc and type def (#2252)
  • @bilgehanertan
    • User API endpoints (#2147)
    • CLI Tag Functionality (#2172)
  • @cjfghk5697
    • ๐ŸŒ [i18n-KO] Translated guides/repository.md to Korean (#2124)
    • ๐ŸŒ [i18n-KO] Translated package_reference/inference_client.md to Korean (#2178)
    • ๐ŸŒ [i18n-KO] Translated package_reference/utilities.md to Korean (#2196)
  • @SeungAhSon
    • ๐ŸŒ [i18n-KO] Translated guides/model_cards.md to Korean" (#2128)
    • ๐ŸŒ [i18n-KO] Translated reference/login.md to Korean (#2151)
    • ๐ŸŒ [i18n-KO] Translated package_reference/hf_file_system.md to Korean (#2174)
  • @seoulsky-field
    • ๐ŸŒ [i18n-KO] Translated guides/community.md to Korean (#2126)
  • @Y4suyuki
    • Define errors in errors.py (#2170)
  • @harheem
    • ๐ŸŒ [i18n-KO] Translated guides/cli.md to Korean (#2131)
    • ๐ŸŒ [i18n-KO] Translated reference/inference_endpoints.md to Korean (#2180)
  • @seoyoung-3060
    • ๐ŸŒ [i18n-KO] Translated guides/search.md to Korean (#2134)
    • ๐ŸŒ [i18n-KO] Translated package_reference/file_download.md to Korean (#2184)
    • ๐ŸŒ [i18n-KO] Translated package_reference/serialization.md to Korean (#2233)
  • @boyunJang
    • ๐ŸŒ [i18n-KO] Translated guides/inference.md to Korean (#2130)
    • ๐ŸŒ [i18n-KO] Translated package_reference/collections.md to Korean (#2214)
    • ๐ŸŒ [i18n-KO] Translated package_reference/space_runtime.md to Korean (#2213)
    • ๐ŸŒ [i18n-KO] Translated guides/manage-spaces.md to Korean (#2220)
  • @nuatmochoi
    • ๐ŸŒ [i18n-KO] Translated guides/webhooks_server.md to Korean (#2145)
    • ๐ŸŒ [i18n-KO] Translated package_reference/cache.md to Korean (#2191)
  • @fabxoe
    • ๐ŸŒ [i18n-KO] Translated package_reference/tensorboard.md to Korean (#2173)
    • ๐ŸŒ [i18n-KO] Translated package_reference/inference_types.md to Korean (#2171)
    • ๐ŸŒ [i18n-KO] Translated package_reference/hf_api.md to Korean (#2165)
    • ๐ŸŒ [i18n-KO] Translated package_reference/mixins.md to Korean (#2166)
  • @junejae
    • ๐ŸŒ [i18n-KO] Translated guides/upload.md to Korean (#2139)
    • ๐ŸŒ [i18n-KO] Translated reference/repository.md to Korean (#2189)
  • @heuristicwave
    • ๐ŸŒ [i18n-KO] Translating guides/hf_file_system.md to Korean (#2146)
  • @usr-bin-ksh
    • ๐ŸŒ [i18n-KO] Translated guides/inference_endpoints.md to Korean (#2164)

so if you want to download a model completely to a local dir with '--local-directory' you don't need to use '--local-dir-use-symlinks False' anymore?
Nevermind, just tested it and now I understand the .huggingface folder. I thought it meant using ~/.cache/huggingface
Thanks for your, and every other contributor's work.

@Anthonyg5005 correct

Sign up or log in to comment