v0.12.0

What's Changed

• 修复bug：请求错误时的报错信息可能泄露apikey by @greenjerry in #39
• 支持模型映射 by @zuisong in #41
• build(deps): bump @hono/node-server from 1.9.1 to 1.10.1 by @dependabot in #42
• 🌱 chore: function call improvement by @zuisong in #47
• Add vercel deploy support by @Radiquum in #48
• New gemini-1.5-flash-latest model by @vuchaev2015 in #49

New Contributors

• @greenjerry made their first contribution in #39
• @Radiquum made their first contribution in #48
• @vuchaev2015 made their first contribution in #49

Full Changelog: v0.11.0...v0.12.0

0.11.0

• Resolve CORS error: Update server configurations to enable Cross-Origin access.
• Add support for function calls in non-stream mode.
• Migrate to itty-router to reduce package size.
• Utilize the official Deno Docker image for improved consistency.

v0.10.0

• gemini-openai-proxy can now act as a reverse proxy for google gemini, which can be useful for people in region that don't have access to the google gemini

  it will request https://generativelanguage.googleapis.com

  curl \
    "http://localhost:8000/v1/models/gemini-pro:generateContent?key=$YOUR_GEMINI_API_KEY" \
  --header 'Content-Type: application/json' \
  --data '{"contents":[{"parts":[{"text":"Hello"}]}]}'

• Service settings can now be passed via apikey, currently the first supported setting is useBeta , which can be set like so

  it will use v1beta version gemini api, https://ai.google.dev/docs/api_versions

  curl http://localhost:8000/v1/chat/completions \
    -H "Authorization: Bearer $YOUR_GEMINI_API_KEY#useBeta" \
    -H "Content-Type: application/json" \
    -d '{
    "model": "gpt-3.5-turbo",
    "messages": [{"role": "user", "content": "Hello"}],
    "temperature": 0.7,"stream":true
    }'

Full Changelog: v0.9.0...v0.10.0