| .. |
|
__tests__
|
168cfcaba5
fix: round-trip Gemini thought signatures for tool calls (#10590)
|
2 weeks ago |
|
cache-strategy
|
62c3914034
Farewell jest (#4607)
|
7 months ago |
|
caching
|
934bfd0a54
feat: Add Vercel AI Gateway provider integration (#7396)
|
5 months ago |
|
anthropic-filter.ts
|
fb9c57e1d9
fix: filter non-Anthropic content blocks before sending to Vertex API (#9618)
|
2 months ago |
|
bedrock-converse-format.ts
|
4693b9eb06
fix(bedrock): convert tool_result to XML text when native tools disabled (#10155)
|
1 month ago |
|
gemini-format.ts
|
168cfcaba5
fix: round-trip Gemini thought signatures for tool calls (#10590)
|
2 weeks ago |
|
image-cleaning.ts
|
5260123689
perf: reduce excessive getModel() calls & implement disk cache fallback (#9410)
|
2 months ago |
|
minimax-format.ts
|
e7c1851a8b
feat(minimax): move environment_details to system message for thinking models (#10284)
|
1 month ago |
|
mistral-format.ts
|
a7b192adca
fix: normalize tool call IDs for cross-provider compatibility via OpenRouter (#10102)
|
1 month ago |
|
model-params.ts
|
de00ab10e2
refactor: consolidate ThinkingBudget components and fix disable handling (#9930)
|
1 month ago |
|
openai-format.ts
|
9ac449919e
fix: ensure assistant message content is never undefined for Gemini compatibility (#10559)
|
2 weeks ago |
|
r1-format.ts
|
970deead5b
fix(deepseek): preserve reasoning_content during tool call sequences (#10141)
|
1 month ago |
|
reasoning.ts
|
048e7f3502
feat(gemini): add minimal and medium reasoning effort levels (#9973)
|
1 month ago |
|
stream.ts
|
0b86796b8f
[feat] Claude Code Provider Native Tool Calling (#10077)
|
1 month ago |
|
vscode-lm-format.ts
|
ade10e275f
fix(vscode-lm): order text parts before tool calls in assistant messages (#10573)
|
2 weeks ago |
|
zai-format.ts
|
eeaf33ce4e
refactor(zai): merge environment_details into tool result instead of system message (#10289)
|
1 month ago |