Commit Graph

171 Commits

Author SHA1 Message Date
daofeng
07cfdaf88a fix(ai-proxy): 处理 Qwen 响应无选择项的情况 (#3448) 2026-02-03 20:33:57 +08:00
澄潭
6c3fd46c6f feat(ai-proxy): add context cleanup command support (#3409) 2026-01-30 17:56:31 +08:00
zikunchang
f2fcd68ef8 feature: Support getting the API key from the request header when provider.apiTokens is not configured. (#3394)
Co-authored-by: 澄潭 <zty98751@alibaba-inc.com>
2026-01-28 14:03:24 +08:00
woody
a2eb599eff Implement Vertex Raw mode support in AI Proxy (#3375) 2026-01-21 14:45:06 +08:00
woody
399d2f372e add support for image generation in Vertex AI provider (#3335) 2026-01-19 16:40:29 +08:00
woody
f1a5f18c78 feat/ai proxy vertex ai compatible (#3324) 2026-01-14 10:13:00 +08:00
woody
23fbe0e9e9 feat(vertex): 为 ai-proxy 插件的 Vertex AI Provider 添加 Express Mode 支持 || feat(vertex): Add Express Mode support to Vertex AI Provider of ai-proxy plug-in (#3301) 2026-01-13 20:00:05 +08:00
CZJCC
78d4b33424 feat(ai-proxy): add Bearer Token authentication support for Bedrock p… (#3305) 2026-01-07 19:39:20 +08:00
rinfx
5cc9f65aaa support disable thinking and add reasoning token usage (#3261) 2025-12-26 17:04:07 +08:00
Bingkun Zhao
e7e3ab5ff6 fix: ai-proxy dify provider extract hostname from difyApiUrl (#3257) 2025-12-24 09:58:54 +08:00
rinfx
e8bcbde5f4 support vertex's claude (#3236) 2025-12-20 10:33:53 +08:00
woody
6998800c64 fix(ai-proxy): ensure basePathHandling works with original protocol (#3225) 2025-12-16 20:49:21 +08:00
rinfx
7f5b37ae6d vertex support global region (#3213) 2025-12-15 17:19:59 +08:00
Liang Deng
5c17d3faa3 feat(ai-proxy): support handle array content in chatToolMessage2BedrockMessage (#3200)
Signed-off-by: Liang Deng <ytdengliang@gmail.com>
Co-authored-by: rinfx <yucheng.lxr@alibaba-inc.com>
2025-12-11 14:15:38 +08:00
rinfx
8a7a375ebd doubao support configuration for domain (#3184) 2025-12-04 21:17:50 +08:00
woody
116e7c6904 implement generic provider for vendor-agnostic passthrough (#3175) 2025-12-03 09:52:47 +08:00
rinfx
7a504fd67d remove omitempty for toolcall index (#3148)
Co-authored-by: 澄潭 <zty98751@alibaba-inc.com>
2025-11-24 16:14:00 +08:00
xujingfeng
c3077d7981 fix(ai-proxy): 调整日志级别以减少冗余警告信息 || fix(ai-proxy): Adjust log level to reduce redundant warning messages (#3120) 2025-11-13 19:24:14 +08:00
woody
8a3c0bb342 feat(ai-proxy): add video-related API paths and capabilities (#3108) 2025-11-11 19:39:49 +08:00
victorserbu2709
1300e09e28 groq add responses capability (#3029)
Co-authored-by: 澄潭 <zty98751@alibaba-inc.com>
2025-11-11 19:36:26 +08:00
rinfx
d4e6704f33 [bugfix] add claude usage & bedrock tool_call index (#3095) 2025-11-10 10:03:01 +08:00
rinfx
1900609fd5 include usage if stream is true (#3084) 2025-11-03 15:36:18 +08:00
woody
5e4c262814 Feat/vllm provider (#3067) 2025-10-29 14:31:38 +08:00
victorserbu2709
5fc1d6b222 Add ApiNameAnthropicMessages in claude capabilities (#3040) 2025-10-23 09:54:47 +08:00
澄潭
271e6036fa feat(ai-proxy): enable Qwen compatible mode by default and add missing API endpoints (#3032) 2025-10-22 11:17:09 +08:00
rinfx
6b2d06a330 [bugfix] sometimes bedrock EventStream chunk is not complete (#3010) 2025-10-16 09:52:51 +08:00
johnlanni
063bfbfcfe fix(ai-proxy): fix streaming process 2025-09-23 19:44:30 +08:00
澄潭
623c8da8d8 fix(ai-proxy): Fix Azure OpenAI Response API handling and service URL type detection (#2948) 2025-09-23 18:49:55 +08:00
aias00
88a679ee07 feat(ai-proxy): add Fireworks AI support (#2917) 2025-09-21 14:32:04 +08:00
rinfx
ef12f40c0e deduplicate think tag for bedrock and vertex (#2933) 2025-09-18 14:35:36 +08:00
rinfx
caae3ee068 [feature] bedrock provider support multimodal and thinking (#2897) 2025-09-18 14:22:37 +08:00
rinfx
d7bebf79e1 vertex support multi-modal, function call and thinking (#2926) 2025-09-18 14:22:22 +08:00
澄潭
e2011cb805 fix(claude): support array content format in tool_result and remove duplicate structs (#2892) 2025-09-10 14:18:44 +08:00
johnlanni
4edf79a1f6 Revert "remove qwen Anthropic compatiable mode"
This reverts commit f7d80373f9.
2025-09-09 21:35:31 +08:00
johnlanni
f7d80373f9 remove qwen Anthropic compatiable mode 2025-09-09 11:12:49 +08:00
aias00
91a44ea7aa feat(provider): add support for meituan longcat || feat(provider): add support for meituan longcat (#2883) 2025-09-08 13:38:40 +08:00
xingpiaoliang
d053e01540 feat(ai-proxy): Add provider: nvidia's triton-server (#2843) 2025-09-08 13:37:30 +08:00
澄潭
4a429bf147 fix(ai-proxy): resolve Claude streaming response conversion and SSE event chunking issues (#2882) 2025-09-08 09:54:18 +08:00
澄潭
44c33617fa feat(ai-proxy): add OpenRouter provider support (#2823) 2025-08-28 19:26:21 +08:00
澄潭
16a18c6609 feat(ai-proxy): add auto protocol compatibility for OpenAI and Claude APIs (#2810) 2025-08-25 14:13:51 +08:00
Xijun Dai
72b98ab6cf feat(ai-proxy): add anthropic/v1/messages and openai/v1/models support for DeepSeek (#2808)
Signed-off-by: Xijun Dai <daixijun1990@gmail.com>
2025-08-21 17:40:13 +08:00
co63oc
eaea782693 fix RegisteTickFunc (#2787) 2025-08-19 20:53:53 +08:00
rinfx
890a802481 feat(ai-proxy): bedrock support tool use (#2730) 2025-08-19 16:54:50 +08:00
Kent Dong
0f1afcdcca fix(ai-proxy): Do not change the configured components of Azure URL (#2782) 2025-08-18 16:27:25 +08:00
Kent Dong
24dca0455e fix: Fix bugs in the bedrock model name escaping logic (#2663)
Co-authored-by: 澄潭 <zty98751@alibaba-inc.com>
2025-08-15 17:40:13 +08:00
xingpiaoliang
0af00bef6b feat(ai-proxy): gemini model multimodal support (#2698) 2025-08-11 15:54:34 +08:00
aias00
a1bf1ff009 feat(provider): add support for Grok provider in AI proxy (#2713)
Co-authored-by: 韩贤涛 <601803023@qq.com>
2025-08-07 17:22:47 +08:00
NOBODY
5ee878198c feat(ai-proxy): gemini model thinking support (#2712) 2025-08-06 06:50:46 +08:00
澄潭
645646fe22 Fix the issue where AI route fallback does not work when using Bedrock. (#2653) 2025-07-31 20:16:16 +08:00
Kent Dong
7348c265b5 feat: Support model mapping and more URL configuration formats for Azure OpenAI (#2649) 2025-07-25 11:28:02 +08:00