Platform for stateful agents: AI with advanced memory that can learn and self-improve over time.
Commits on main
feat: Add toggle on llm config for parallel tool calling [LET-5610] (#5542)
396959d Matthew Zhou committed 1 year ago
feat: inject `letta_client` and `agent_id` into local sandbox (#5192)
305bb8c Sarah Wooders committed 1 year ago
feat: add comprehensive testing for client side tool calling (#5539)
5a475fd cthomas committed 1 year ago
feat: support approval requests for parallel tool calls (#5538)
69b15d6 cthomas committed 1 year ago
fix: ensure messages are in chronological order when backfilling tool call id (#5520)
5913730 jnjpng committed 1 year ago
feat: handle noop early return case first in agent loop (#5536)
181a73c cthomas committed 1 year ago
feat: rename is_approval var to is_approval_response for clarity (#5534)
507a83a cthomas committed 1 year ago
feat: add validation to fastapi routes for agent IDs (#5454)
4bbd760 Kian Jones committed 1 year ago
providers changes, no changes to embeddings or llms [LET-4630] (#5474)
c672a1c Ari Webb committed 1 year ago
test: Swap to letta agent v1 for agent serialization (#5464)
fc32db3 Matthew Zhou committed 1 year ago
feat: Implement streaming chat completions endpoint [LET-5485] (#5446)
2dae4d3 Matthew Zhou committed 1 year ago