Headline
- Second beta release for Lemonade C++
What's Changed
C++
- C++: Enable llamacpp VLMs by @jeremyfowers in #532
- C++: linux CLI fixes and new tests by @jeremyfowers in #534
- C++: Fix false DLL dependence by @jeremyfowers in #535
- C++: Improve model endpoint performance by @jeremyfowers in #541
- C++: Add support for HF_HUB_CACHE by @jeremyfowers in #542
- C++: Enable stats endpoint, add prompt_tokens by @jeremyfowers in #543
Python
- Fix dll issue in RAI 1.6 by @ramkrishna2910 in #530
- Fix FLM Chat Completions by @itomek in #481
- Added a missing --device support for llamacpp by @meghsat in #525
Both C++ and Python
- Add enable_thinking toggle to web UI model settings by @Kritik-07 in #531
- Simplify and fix the web ui by @jeremyfowers in #544
Full Changelog: v8.2.1...v8.2.2