Release 0.35.8

Released 2021-11-07.

• Fixed token counting for long prompts
• Added per-model rate limits
• Faster model list caching

Release 48417 · Release 48418 · Release 70229

Back to the changelog