jnjpng
5017cb1d12
feat: add chatgpt oauth client for codex routing ( #8774 )
...
* base
* refresh
* use default model fallback
* patch
* streaming
* generate
2026-01-19 15:54:42 -08:00
Ari Webb
0a372b2540
fix: enable zai streaming ( #7755 )
2026-01-12 10:57:20 -08:00
Charles Packer
33d39f4643
fix(core): patch usage data tracking for anthropic when context caching is on ( #6997 )
2025-12-15 12:03:09 -08:00
Devansh Jain
d1536df6f6
chore: Update deepseek client for v3.2 models ( #6556 )
...
* support for v3.2 models
* streaming + context window fix
* fix for no assitant text from deepseek
2025-12-15 12:02:34 -08:00
Kian Jones
edeac2c679
fix: fix gemini otel bug and add tracing for tool upsert ( #6523 )
...
add tracing for tool upsert, and fix gemini otel bug
2025-12-15 12:02:33 -08:00
Kian Jones
a38475f23d
fix: safely load span attributes for provider traces ( #6508 )
...
json.dumps on request data. Also remove step and actor since they are already present in the span
2025-12-15 12:02:33 -08:00
Kian Jones
5165d60881
feat: add a new span and log the provider request and response data objects ( #6492 )
...
add a new span and log the provider request and response data objects
2025-12-15 12:02:33 -08:00
Charles Packer
1f7165afc4
fix: patch counting of tokens for anthropic ( #6458 )
...
* fix: patch counting of tokens for anthropic
* fix: patch ui to be simpler
* fix: patch undercounting bug in anthropic when caching is on
2025-12-15 12:02:19 -08:00
Charles Packer
e67c98eedb
feat: add tests for prompt caching + fix anthropic prompt caching [LET-6373] ( #6454 )
...
* feat: add tests for prompt caching
* fix: add cache control breakpoints for anthropic + fix tests
* fix: silence logging
* fix: patch token counting error
* fix: same patch on non-streaming path
2025-12-15 12:02:19 -08:00
Charles Packer
4af6465226
feat(core+web): store raw usage data on streams (and visualize properly in ADE) ( #6452 )
...
* feat(core): store raw usage data on streams
* fix(web): various fixes to deal w/ hardcoding against openai
2025-12-15 12:02:19 -08:00
Charles Packer
88a3743cc8
fix(core): distinguish between null and 0 for prompt caching ( #6451 )
...
* fix(core): distinguish between null and 0 for prompt caching
* fix: runtime errors
* fix: just publish just sgate
2025-12-15 12:02:19 -08:00
Charles Packer
131891e05f
feat: add tracking of advanced usage data (eg caching) [LET-6372] ( #6449 )
...
* feat: init refactor
* feat: add helper code
* fix: missing file + test
* fix: just state/publish api
2025-12-15 12:02:19 -08:00
jnjpng
c6df306ccf
fix: upgrade google-genai sdk version and fix gemini 3 streaming ( #6437 )
...
* base
* base
---------
Co-authored-by: Letta Bot <noreply@letta.com >
2025-12-15 12:02:18 -08:00
Ari Webb
30dab0abb9
fix: handle llm error during streaming [LET-6280] ( #6341 )
...
handle llm error during streaming
Co-authored-by: Ari Webb <ari@letta.com >
2025-11-24 19:10:27 -08:00
Charles Packer
18029250d0
fix(core): sanitize messages to anthropic in the main path the same way (or similar) to how we do it in the token counter ( #6044 )
...
* fix(core): sanitize messages to anthropic in the main path the same way (or similar) to how we do it in the token counter
* fix: also patch poison error in backend by filtering lazily
* fix: remap streaming errors (what the fuck)
* fix: dedupe tool clals
* fix: cleanup, removed try/catch
2025-11-13 15:36:55 -08:00
Matthew Zhou
ff81f4153b
feat: Support parallel tool calling streaming for OpenAI chat completions [LET-4594] ( #5865 )
...
* Finish chat completions parallel tool calling
* Undo comments
* Add comments
* Remove test file
2025-11-13 15:36:14 -08:00
Ari Webb
48cc73175b
feat: parallel tool calling for openai non streaming [LET-4593] ( #5773 )
...
* first hack
* clean up
* first implementation working
* revert package-lock
* remove openai test
* error throw
* typo
* Update integration_test_send_message_v2.py
* Update integration_test_send_message_v2.py
* refine test
* Only make changes for openai non streaming
* Add tests
---------
Co-authored-by: Ari Webb <ari@letta.com >
Co-authored-by: Matt Zhou <mattzh1314@gmail.com >
2025-11-13 15:36:14 -08:00
Matthew Zhou
bb8a7889e0
feat: Add parallel tool call streaming for anthropic [LET-4601] ( #5225 )
...
* wip
* Fix parallel tool calling interface
* wip
* wip adapt using id field
* Integrate new multi tool return schemas into parallel tool calling
* Remove example script
* Reset changes to llm stream adapter since old agent loop should not enable parallel tool calling
* Clean up fallback logic for extracting tool calls
* Remove redundant check
* Simplify logic
* Clean up logic in handle ai response
* Fix tests
* Write anthropic dict conversion to be back compatible
* wip
* Double write tool call id for legacy reasons
* Fix override args failures
* Patch for approvals
* Revert comments
* Remove extraneous prints
2025-10-24 15:11:31 -07:00
Matthew Zhou
7511b0f4fe
feat: Write anthropic streaming interface that supports parallel tool calling [LET-5355] ( #5295 )
...
Write anthropic streaming interface that supports parallel tool calling
2025-10-09 15:25:21 -07:00
cthomas
1d611d92b9
feat: update assistant content parts union ( #5115 )
...
* feat: update assistant content parts union
* api sync
* just use the base object since updating assistant breaks frontend
2025-10-07 17:50:48 -07:00
cthomas
f7755d837a
feat: add gemini streaming to new agent loop ( #5109 )
...
* feat: add gemini streaming to new agent loop
* add google as required dependency
* support storing all content parts
* remove extra google references
2025-10-07 17:50:48 -07:00
Sarah Wooders
ef07e03ee3
feat: add run_id to input messages and step_id to messages ( #5099 )
2025-10-07 17:50:48 -07:00
cthomas
67f8e46619
feat: add run id to streamed messages ( #5037 )
2025-10-07 17:50:47 -07:00
Matthew Zhou
d3c5d0c330
feat: Add missing import for SimpleOpenAIResponsesStreamingInterface ( #5036 )
...
Add missing import
2025-10-07 17:50:47 -07:00
cthomas
76d1bc8cbc
feat: move new streaming adapters into own files ( #5001 )
2025-10-07 17:50:47 -07:00
