v0.6.6

vllm-project/vllm-omniv0.6.6May 27, 2024by github-actions[bot]

AI Summary

This release adds support for new AI providers including AWS Bedrock, Groq, and Cohere, alongside updates to Gemini and Ollama integrations. It also includes various bug fixes for logger races, Claude streaming responses, and dependency updates.

Key Highlights

Support for AWS Bedrock runtime Claude 3
Groq support for Llama 3
Cohere integration added
Gemini function and tools support
Ollama vision model support

New Features

AWS Bedrock runtime Claude 3 support
Groq Llama 3 support
Cohere integration
Gemini function and tools support
Ollama vision model support
Dependency bump golang.org/x/net

Full Release Notes

## What's Changed
* chore(deps): bump golang.org/x/net from 0.17.0 to 0.23.0 by @dependabot in https://github.com/songquanpeng/one-api/pull/1335
* feat: support aws bedrockruntime claude3 by @Laisky in https://github.com/songquanpeng/one-api/pull/1328
* fix: logger race by @Laisky in https://github.com/songquanpeng/one-api/pull/1339
* feat: groq support Llama3 now by @tylinux in https://github.com/songquanpeng/one-api/pull/1333
* feat: add function and tools support for Gemini by @mxdlzg in https://github.com/songquanpeng/one-api/pull/1358
* fix: fix default theme blank screen when edit channel again by @tylinux in https://github.com/songquanpeng/one-api/pull/1363
* feat: add cohere support by @ye4293 in https://github.com/songquanpeng/one-api/pull/1355
* fix: refactor Gemini adaptor to support streaming content generation by @mxdlzg in https://github.com/songquanpeng/one-api/pull/1382
* feat: support for the ollama vision model by @nongmo677 in https://github.com/songquanpeng/one-api/pull/1376
* feat: use mapped model name to test by @tylinux in https://github.com/songquanpeng/one-api/pull/1370
* fix: claude stream response parse by @WqyJh in https://github.com/songquanpeng/one-api/pull/1334
* fix: fix zhipu embedding error when input is array but not string by @caixinjiang in https://github.com/songquanpeng/one-api/pull/1306
* fix: fix getPreConsumedQuota by @plusye in https://github.com/songquanpeng/one-api/pull/1312

## New Contributors
* @tylinux made their first contribution in https://github.com/songquanpeng/one-api/pull/1333
* @mxdlzg made their first contribution in https://github.com/songquanpeng/one-api/pull/1358
* @nongmo677 made their first contribution in https://github.com/songquanpeng/one-api/pull/1376
* @caixinjiang made their first contribution in https://github.com/songquanpeng/one-api/pull/1306
* @plusye made their first contribution in https://github.com/songquanpeng/one-api/pull/1312

**Full Changelog**: https://github.com/songquanpeng/one-api/compare/v0.6.5...v0.6.6