| .. |
|
ai360
|
192fc094fc
fix: 360智能接口地址更新
|
hai 1 ano |
|
ali
|
a972722367
fix: 使用openai兼容接口调用部分渠道在最终端点为claude原生端点下还是走了openai扣减input_token的逻辑
|
hai 4 meses |
|
aws
|
e194b747c3
feat: support adaptive thinking
|
hai 4 meses |
|
baidu
|
c2ed76ddfd
🛡️ fix: prevent OOM on large/decompressed requests; skip heavy prompt meta when token count is disabled
|
hai 6 meses |
|
baidu_v2
|
d98e7b3dbf
Merge pull request #2296 from seefs001/fix/adapter-missing
|
hai 6 meses |
|
claude
|
c494c26236
Merge pull request #2887 from seefs001/fix/claude
|
hai 4 meses |
|
cloudflare
|
1fededceb3
feat: refactor token estimation logic
|
hai 6 meses |
|
codex
|
47cdac3e64
Merge pull request #2883 from seefs001/fix/claude-relay-info-input-token
|
hai 4 meses |
|
cohere
|
1fededceb3
feat: refactor token estimation logic
|
hai 6 meses |
|
coze
|
c2ed76ddfd
🛡️ fix: prevent OOM on large/decompressed requests; skip heavy prompt meta when token count is disabled
|
hai 6 meses |
|
deepseek
|
a972722367
fix: 使用openai兼容接口调用部分渠道在最终端点为claude原生端点下还是走了openai扣减input_token的逻辑
|
hai 4 meses |
|
dify
|
1fededceb3
feat: refactor token estimation logic
|
hai 6 meses |
|
gemini
|
99928bcfde
fix: charge local input tokens when Gemini returns empty response
|
hai 4 meses |
|
jimeng
|
5010f2d004
format: package name -> github.com/QuantumNous/new-api (#2017)
|
hai 8 meses |
|
jina
|
5010f2d004
format: package name -> github.com/QuantumNous/new-api (#2017)
|
hai 8 meses |
|
lingyiwanwu
|
3f6aeaa0e3
Update Perplexity and 01 models
|
%!s(int64=2) %!d(string=hai) anos |
|
minimax
|
ffa8a42784
fix(minimax): 添加 MiniMax-M2 系列模型到 ModelList
|
hai 5 meses |
|
mistral
|
5010f2d004
format: package name -> github.com/QuantumNous/new-api (#2017)
|
hai 8 meses |
|
mokaai
|
5010f2d004
format: package name -> github.com/QuantumNous/new-api (#2017)
|
hai 8 meses |
|
moonshot
|
a972722367
fix: 使用openai兼容接口调用部分渠道在最终端点为claude原生端点下还是走了openai扣减input_token的逻辑
|
hai 4 meses |
|
ollama
|
9ef9e78821
feat(file): unify file handling with a new FileSource abstraction for URL and base64 data
|
hai 4 meses |
|
openai
|
3d0c75f42c
Merge branch 'feature/messages2responses' into upstream-main
|
hai 4 meses |
|
openrouter
|
6682842cbb
feat: enhance OpenRouter enterprise support with new settings and response handling
|
hai 8 meses |
|
palm
|
1fededceb3
feat: refactor token estimation logic
|
hai 6 meses |
|
perplexity
|
b587796929
feat: /v1/responses perplexity
|
hai 4 meses |
|
replicate
|
af671d34c3
feat: replicate channel flux model
|
hai 7 meses |
|
siliconflow
|
5d12755798
refactor(relay): enhance error logging and improve multipart form handling in audio requests #2127
|
hai 7 meses |
|
submodel
|
5010f2d004
format: package name -> github.com/QuantumNous/new-api (#2017)
|
hai 8 meses |
|
task
|
0823c16f97
feat: doubao add first and last image to video
|
hai 4 meses |
|
tencent
|
1fededceb3
feat: refactor token estimation logic
|
hai 6 meses |
|
vertex
|
a972722367
fix: 使用openai兼容接口调用部分渠道在最终端点为claude原生端点下还是走了openai扣减input_token的逻辑
|
hai 4 meses |
|
volcengine
|
a972722367
fix: 使用openai兼容接口调用部分渠道在最终端点为claude原生端点下还是走了openai扣减input_token的逻辑
|
hai 4 meses |
|
xai
|
1fededceb3
feat: refactor token estimation logic
|
hai 6 meses |
|
xinference
|
038202b1f6
🔧 fix(xinference): update Document type to 'any' for flexibility
|
hai 11 meses |
|
xunfei
|
5010f2d004
format: package name -> github.com/QuantumNous/new-api (#2017)
|
hai 8 meses |
|
zhipu
|
5010f2d004
format: package name -> github.com/QuantumNous/new-api (#2017)
|
hai 8 meses |
|
zhipu_4v
|
a972722367
fix: 使用openai兼容接口调用部分渠道在最终端点为claude原生端点下还是走了openai扣减input_token的逻辑
|
hai 4 meses |
|
adapter.go
|
920e005048
fix: Use channel proxy settings for task query scenarios
|
hai 6 meses |
|
api_request.go
|
0e095d4ad8
feat(api): add 'cookie' to passthroughSkipHeaderNamesLower
|
hai 4 meses |
Calcium-Ion
6c28de8965
Merge pull request #2889 from seefs001/feature/messages2responses