text-generation-inference
feat: allow tool calling to respond without a tool
#2614
Merged

Commits
  • feat: process token stream before returning to client
    drbh committed 1 year ago
  • fix: expect content in test
    drbh committed 1 year ago
  • fix: improve comparison via ruff lint
    drbh committed 1 year ago
  • fix: return event in all cases
    drbh committed 1 year ago
  • fix: always send event on error, avoid unwraps, refactor and improve tests
    drbh committed 1 year ago
  • fix: prefer no_tool over notify_error to improve reponse
    drbh committed 1 year ago
  • fix: adjust chat input test for no_tool
    drbh committed 1 year ago
  • fix: adjust test expected content
    drbh committed 1 year ago
Loading