Skip to content

Releases: friendliai/friendli-client

Release v1.4.1 🚀

19 Jun 06:13
Compare
Choose a tag to compare

Updating Patch Version

This patch version Introduces explicit resource management to prevent unexpected resource leaks.
By default, the library closes underlying HTTP and gRPC connections when the client is garbage-collected. However, you can now manually close the Friendli or AsyncFriendli client using the .close() method or utilize a context manager to ensure proper closure when exiting a with block.

Usage examples

import asyncio
from friendli import AsyncFriendli

client = AsyncFriendli(base_url="0.0.0.0:8000", use_grpc=True)

async def run():
    async with client:
        stream = await client.completions.create(
            prompt="Explain what gRPC is. Also give me a Python code snippet of gRPC client.",
            stream=True,
            top_k=1,
        )

        async for chunk in stream:
            print(chunk.text, end="", flush=True)

asyncio.run(run())

Release v1.4.0 🚀

18 Jun 06:45
Compare
Choose a tag to compare
  • gRPC client support for completions API.

Release v1.3.7 🚀

12 Jun 03:10
Compare
Choose a tag to compare
  • Minor: add a default value for the "index" and "text" fields of the completion stream's chunk.

Release v1.3.6 🚀

10 Jun 06:40
Compare
Choose a tag to compare
  • Support Phi3 FP8 conversion.
  • Hotfix for safetensor checkpoint saver.

Release v1.3.5 🚀

25 May 04:11
Compare
Choose a tag to compare
  • Optimize CPU RAM usage during quantization with offloading
  • Support FP8 conversion for DBRX, Mixtral, and Command R+

Release v1.3.4 🚀

02 Apr 03:22
Compare
Choose a tag to compare
  • Hotfix for LoRA checkpoint saving error.

Release v1.3.3 🚀

01 Apr 05:35
Compare
Choose a tag to compare

New Features

  • FP8 Checkpoint Conversion: We've introduced a new feature for FP8 checkpoint conversion.
  • Sharded Safetensors Checkpoint Saving: Added the ability to save sharded safetensors checkpoints.
  • LoRA Support on Mistral Model: We have added support for LoRA (Low-Rank Adaptation) on the Mistral model.

Bug Fixes

  • BF16 Hotfix: Addressed an urgent issue with bf16 processing.
  • BFloat Safetensors Conversion: Fixed an issue related to bfloat conversion for safetensors.
  • Automatic Token Refresh: Resolved a bug affecting automatic token refresh.

Release v1.3.2 🚀

26 Mar 14:45
Compare
Choose a tag to compare
  • Add base_model_name_or_path option to friendli model convert-adapter.
  • Remove stale dependencies.

Release v1.3.1 🚀

25 Mar 10:54
Compare
Choose a tag to compare
  • Update protobuf schema.
  • Patch sending API requests with content type application/protobuf.

Release v1.3.0 🚀

23 Mar 16:10
Compare
Choose a tag to compare
  • Now resources of Friendli Dedicated Endpoints can be managed with CLI and SDK. The available resources are endpoint, model, team, and project.
  • Login with CLI is now available. SSO login is also available.
  • Update on Multi-LoRA checkpoint conversion.