Code

History

Commits on fix-git-https-support to crates/collab/src/llm.rs70+
  1. Remove Qwen2 model (#18444)Richard Feldman committed 2 years agocaaa9a0
  2. remote_server: Remove dependency on libssl and libcrypto (#15446)Piotr Osiewicz committed 2 years ago2c8a6ee
  3. /auto (#16696)Richard Feldman committed 2 years ago91ffa02
  4. chore: Fix several style lints (#17488)Piotr Osiewicz committed 2 years agoe6c1c51
  5. assistant: Fix Google AI provider not respecting `low_speed_timeout_in_seconds` (#17423)Bennet Bo Fenner committed 2 years agof413ea9
  6. collab: Add `GET /models` endpoint to LLM service (#17307)Marshall Bowers committed 2 years ago3005625
  7. collab: Report when upstream rate limit is exceeded (#17083)Marshall Bowers committed 2 years agod666cc5
  8. collab: Add lifetime spending limit for LLM usage (#16780)Marshall Bowers committed 2 years ago340662e
  9. collab: Report active user counts separately, as well (#16629)Marshall Bowers committed 2 years ago1d986b0
  10. collab: Track active user counts independently for each model (#16624)Marshall Bowers committed 2 years ago0229d3c
  11. collab: Add traces for user LLM rate limits (#16610)Marshall Bowers committed 2 years ago96bccee
  12. collab: Add `is_staff` to upstream rate limit spans (#16463)Marshall Bowers committed 2 years agode41c15
  13. collab: Add `is_staff` to spans (#16389)Marshall Bowers committed 2 years ago3d997e5
  14. Add tracing needed for LLM rate limit dashboards (#16388)Max Brunsfeld committed 2 years ago1b1070e
  15. collab: Rework model name checks (#16365)Marshall Bowers committed 2 years ago7a5acc0
  16. collab: Attach GitHub login to LLM spans (#16316)Marshall Bowers committed 2 years ago9233418
  17. collab: Attach `user_id` to LLM spans (#16311)Marshall Bowers committed 2 years ago5e05821
  18. collab: Add ability to revoke LLM service access tokens (#16143)Marshall Bowers committed 2 years agob4c22cc
  19. Track lifetime spending for each user and model (#16137)Max Brunsfeld committed 2 years agodbcd066
  20. Report telemetry events for rate limit errors (#16130)Max Brunsfeld committed 2 years agoa3c7921
  21. Expose anthropic API errors to the client (#16129)Max Brunsfeld committed 2 years ago1674e12
  22. collab: Use a separate Anthropic API key for Zed staff (#16128)Marshall Bowers committed 2 years agof3ec8d4
  23. Surface upstream rate limits from Anthropic (#16118)Marshall Bowers committed 2 years agoebdb755
  24. collab: Temporarily bypass LLM rate limiting for staff (#16089)Marshall Bowers committed 2 years ago3140d6c
  25. Capture telemetry data on per-user monthly LLM spending (#16050)Max Brunsfeld committed 2 years ago33e120d
  26. Add telemetry for LLM usage (#16049)Max Brunsfeld committed 2 years ago8688b2a
  27. Use LLM service for tool call requests (#16046)Max Brunsfeld committed 2 years agofbebb73
  28. Fix usage recording in llm service (#16044)Max Brunsfeld committed 2 years agob1c69c2
  29. Remove code paths that skip LLM db in prod (#16008)Max Brunsfeld committed 2 years ago225726b
  30. Apply rate limits in LLM service (#15997)Max Brunsfeld committed 2 years ago06625bf
  31. collab: Always use newest anthropic model version (#15978)Bennet Bo Fenner committed 2 years ago514b79e
  32. collab: Setup database for LLM service (#15882)Marshall Bowers committed 2 years ago7f6d091
  33. Authorize access to language model providers based on country (#15859)Marshall Bowers committed 2 years agocf5f4dd
  34. collab: Add support for more providers to the LLM service (#15832)Marshall Bowers committed 2 years agoca95113
  35. Introduce a separate backend service for LLM calls (#15831)Max Brunsfeld committed 2 years ago8e9c2b1