This issue serves as a "news ticker" for the Aviary frontend.
Current news:
We're sunsetting the non-Llama models as of the 0.3.0 release. The reason is because we've seen the demand for llama models significantly outpace the Falcon and MPT models along with corresponding accuracy improvements.
Please feel free to create an issue if you would like to see new issues.
Past updates:
[2023-08-28] Sunsetting non-Llama model examples on RayLLM - see chat.lmsys.org for others!
[2023-07-20] Just added: Llama 2 models All Sizes!!!
[2023-07-03] Just added: continuous batching! Refreshed model list!
[2023-06-22] Just added: mpt-30b-chat! Streaming support & other improvements!
[2023-06-21] Just added: Streaming support & other improvements!
[2023-06-15] Falcon Models have a bug that slows them down and cause timeouts.
[2023-06-06] Just added: Falcon models! KubeRay guide!
[2023-06-02] Just added: Expanded context windows (900 words)! OpenAI CLI support! Upcoming: Falcon 40b
This issue serves as a "news ticker" for the Aviary frontend.
Current news:
We're sunsetting the non-Llama models as of the 0.3.0 release. The reason is because we've seen the demand for llama models significantly outpace the Falcon and MPT models along with corresponding accuracy improvements.
Please feel free to create an issue if you would like to see new issues.
Past updates: