⭐ GitHub Trending Daily RAG

精选免费 AI API、模型、基础设施与开发框架,持续更新。

共 10 个资源 · 更新于 2026-07-14 20:00:00 UTC

🔥 热门

Google AI Studio 提供 Gemini 1.5 Flash / Pro 免费调用,每分钟 15 次请求,适合个人开发与原型验证。

免费额度

免费层 15 RPM / 1500 RPD,上下文 100 万 tokens

大模型 API 更新于 2026-07-10
#gemini #llm #free-tier #api
🔥 热门

在 Cloudflare 边缘网络运行 Llama、Mistral、Whisper 等开源模型,每天 10000 次免费神经元推理。

免费额度

每天 10000 次神经元推理免费

边缘推理 更新于 2026-07-12
#cloudflare #edge #llama #inference
🔥 热门

Groq 使用 LPU 硬件,Llama 3 / Mixtral 推理速度可达每秒 500+ tokens,免费 API 适合实时场景。

免费额度

免费层 30 RPM,日上限 14400 tokens

大模型 API 更新于 2026-07-11
#groq #llama #fast-inference
🔥 热门

GitHub 为所有用户开放免费版 Copilot,每月 2000 次代码补全、50 次聊天,VS Code / JetBrains 全 IDE 支持。

免费额度

每月 2000 次补全 + 50 次聊天

编程助手 更新于 2026-07-09
#copilot #coding #ide
✨ 推荐

HF Serverless Inference API 提供数千个开源模型的免费调用,适合实验、原型与教学。

免费额度

免费层有限调用,需登录

开源模型 更新于 2026-07-08
#huggingface #open-source #models
✨ 推荐

OpenRouter 聚合多家供应商,标记 `:free` 的模型完全免费,统一 OpenAI 兼容接口,适合多模型对比。

免费额度

多个 `:free` 模型,每天 50 次请求

大模型 API 更新于 2026-07-13
#openrouter #aggregator #free-models
✨ 推荐

Supabase 提供免费 PostgreSQL + pgvector,500MB 数据库 + 5GB 带宽,适合构建 RAG 应用。

免费额度

免费层 500MB 数据库 + 50000 月活用户

基础设施 更新于 2026-07-07
#supabase #pgvector #rag #postgres
✨ 推荐

Together AI 新用户赠送 $5 信用,可调用 Llama、Qwen、DeepSeek 等开源模型,推理稳定且支持微调。

免费额度

新用户 $5 免费额度

大模型 API 更新于 2026-07-06
#together #open-source #credits
✨ 推荐

阿里魔搭社区提供通义千问、百川等开源模型的免费在线推理与微调,中文场景表现优秀。

免费额度

免费在线推理 + CPU 计算资源

开源模型 更新于 2026-07-05
#modelscope #qwen #chinese
📦 收录

Vercel 开源 AI SDK,统一接口对接 OpenAI / Anthropic / Google 等,hobby 计划每月 1000 次免费函数调用。

免费额度

hobby 计划 1000 次 / 月 函数调用

开发框架 更新于 2026-07-04
#vercel #ai-sdk #framework