| .. |
|
ai360
|
192fc094fc
fix: 360智能接口地址更新
|
hai 1 ano |
|
ali
|
5d7e688bfe
Merge pull request #1174 from QuantumNous/refactor/message
|
hai 1 ano |
|
aws
|
0a7842e897
fix aws claude-sonnet-4-20250514
|
hai 1 ano |
|
baidu
|
8343fe0740
refactor: message content 改成 any
|
hai 1 ano |
|
baidu_v2
|
46f95fddcd
feat: add ToMap method and enhance OpenAI request handling
|
hai 1 ano |
|
claude
|
3164e86278
fix: remove unnecessary error handling in token counting functions
|
hai 1 ano |
|
cloudflare
|
16997a695d
refactor: token counter logic
|
hai 1 ano |
|
cohere
|
16997a695d
refactor: token counter logic
|
hai 1 ano |
|
coze
|
3164e86278
fix: remove unnecessary error handling in token counting functions
|
hai 1 ano |
|
deepseek
|
8ea4c76f2b
feat: support /v1/responses API
|
hai 1 ano |
|
dify
|
3164e86278
fix: remove unnecessary error handling in token counting functions
|
hai 1 ano |
|
gemini
|
d244915111
✨ feat(relay-gemini): conditionally set ThinkingBudget based on MaxOutputTokens
|
hai 1 ano |
|
jina
|
8ea4c76f2b
feat: support /v1/responses API
|
hai 1 ano |
|
lingyiwanwu
|
3f6aeaa0e3
Update Perplexity and 01 models
|
%!s(int64=2) %!d(string=hai) anos |
|
minimax
|
3aae4efe8c
feat: support minimax
|
%!s(int64=2) %!d(string=hai) anos |
|
mistral
|
07a671ec27
🐛 fix: update condition to check for empty content in assistant role messages
|
hai 1 ano |
|
mokaai
|
8ea4c76f2b
feat: support /v1/responses API
|
hai 1 ano |
|
moonshot
|
3aae4efe8c
feat: support minimax
|
%!s(int64=2) %!d(string=hai) anos |
|
ollama
|
8ea4c76f2b
feat: support /v1/responses API
|
hai 1 ano |
|
openai
|
16997a695d
refactor: token counter logic
|
hai 1 ano |
|
openrouter
|
a813b24184
feat: support claude cache and thinking for upstream [OpenRouter] (#983)
|
hai 1 ano |
|
palm
|
16997a695d
refactor: token counter logic
|
hai 1 ano |
|
perplexity
|
8ea4c76f2b
feat: support /v1/responses API
|
hai 1 ano |
|
siliconflow
|
8ea4c76f2b
feat: support /v1/responses API
|
hai 1 ano |
|
task
|
aaf7c88e0b
feat: add video channel kling fix
|
hai 1 ano |
|
tencent
|
16997a695d
refactor: token counter logic
|
hai 1 ano |
|
vertex
|
4f84fd6a71
fix: update model name logic for vertex
|
hai 1 ano |
|
volcengine
|
ce21b247bd
feat: 火山引擎增加文生图
|
hai 1 ano |
|
xai
|
16997a695d
refactor: token counter logic
|
hai 1 ano |
|
xinference
|
8918381c96
feat: support xinference rerank to jina format
|
hai 1 ano |
|
xunfei
|
8343fe0740
refactor: message content 改成 any
|
hai 1 ano |
|
zhipu
|
8343fe0740
refactor: message content 改成 any
|
hai 1 ano |
|
zhipu_4v
|
8ea4c76f2b
feat: support /v1/responses API
|
hai 1 ano |
|
adapter.go
|
d7ed0214ad
feat: add video channel kling
|
hai 1 ano |
|
api_request.go
|
92de4d0d04
🔧 fix(api_request): enhance ping keep-alive mechanism with error handling and timeout controls
|
hai 1 ano |
CaIon
d244915111
✨ feat(relay-gemini): conditionally set ThinkingBudget based on MaxOutputTokens