Star 历史趋势
数据来源: GitHub API · 生成自 Stargazers.cn
README.md

awesome-free-llm-apis

342+ free LLM APIs from 30 providers — find, compare & configure free models in seconds.

🌐 Live at freellm.netBrowse models · Playground · Config generator · API keys

Provider Logos

🔄 Data refreshed daily from freellm.net — Last updated: 2026-07-07

🌐 English · 简体中文 · 繁體中文 · 日本語 · 한국어


Why This Exists

Finding a free LLM API shouldn't mean hunting through a dozen GitHub READMEs, signing up for five different platforms, or guessing which models still have a free tier.

This repo is a structured, machine-readable directory of every free LLM API — rate limits, context windows, one-click config snippets, and direct API key links. Updated daily.

Why this repo + freellm.net:

  • Always up-to-date — data refreshed daily via automated monitoring, not a 2-year-old static list
  • Credit card transparency — clearly shows which providers require a card, phone verification, or nothing at all
  • One-click configs — ready-to-copy snippets for Claude Code, Cursor, Codex, Aider, and 10+ more tools
  • Side-by-side comparison — compare context windows, rate limits, and modalities across providers instantly

How to Use — 3 Steps

  1. Pick a provider — see Provider Directory below. Start with Groq (no credit card, 30 RPM free).
  2. Get your API key — click any Get Key → link below, sign up (most need just an email), and copy your key. Takes < 1 minute.
  3. Plug it in — copy the base URL + model ID, paste into the Quick Start examples below.

Configuring a specific tool? Claude Code · Cursor · Codex · OpenHuman · OpenCode · OpenClaw — one-click configs at freellm.net/config/.

Quick Start — Use Any Free API in 30 Seconds

Never used an API before? Here's the simplest path: go to console.groq.com/keys, sign up with just an email (no credit card), copy your free key, and paste it into any example below. You'll be running in under a minute.

All providers below expose an OpenAI-compatible endpoint. Any tool that accepts baseURL + apiKey works — just swap the base URL and key.

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://api.groq.com/openai/v1",  # free, no credit card
    api_key="GROQ_API_KEY",                     # get at console.groq.com/keys
)

response = client.chat.completions.create(
    model="llama-3.3-70b-versatile",            # see Best Models table below
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
# Groq free tier: 30 RPM, 14,400 RPD — generous for personal use

Codex CLI

export OPENAI_BASE_URL="https://api.groq.com/openai/v1"
export OPENAI_API_KEY="your-groq-key"          # get at console.groq.com/keys
codex --model "llama-3.3-70b-versatile"

Cursor

Settings → Models → Add Model
  Model name: llama-3.3-70b-versatile
  Base URL: https://api.groq.com/openai/v1
  API key: your-groq-key                       # get at console.groq.com/keys

Claude Code

# Claude Code needs an Anthropic-compatible API — use OpenRouter
export ANTHROPIC_BASE_URL="https://openrouter.ai/api"
export ANTHROPIC_AUTH_TOKEN="sk-or-v1-your-key"  # openrouter.ai/keys
export ANTHROPIC_API_KEY=""                       # must be empty
# Note: OpenRouter Anthropic models need $10 top-up (one-time)

Using Other Tools?

Most AI dev tools accept custom API endpoints — just point them at any provider above. Grab your free key, then:

More ready-to-copy configs at freellm.net/config/.

All providers, base URLs, and API key links are in the Quick Reference below.


Provider Directory

⚡ Permanent Free Tiers

These providers offer a permanently free tier — no credit card required for most.

ProviderFree ModelsCredit Card?Max ContextModalitiesGet API Key
NVIDIA NIM104Phone verification1Maudio, embedding, image, reasoning, rerank, text, video
Cloudflare Workers AI38No10Mcode, image, reasoning, text, video
ModelScope33Registration1Maudio, image, reasoning, text, video, vision
GitHub Models13No1Maudio, image, pdf, reasoning, text
OVHcloud AI Endpoints12Registration262Kaudio, code, image, reasoning, text, video
Groq11No262Kimage, reasoning, text
Google Gemini10No2Maudio, image, pdf, reasoning, text, video, vision
Mistral AI9No256Kcode, image, text
LLM7.io8No256Kaudio, code, image, pdf, reasoning, text, video
Cerebras6No131Kreasoning, text
Cohere6No256Ktext
Ollama Cloud6Registration262Kcode, reasoning, text
Agnes AI5Registration256Kimage, text, video, vision
Aion Labs5Registration131Ktext
Hugging Face5No131Ktext
Kilo Code5No262Kcode, reasoning, text
Alibaba Cloud Model Studio5Registration1Mcode, image, text
OpenCode Zen5Registration1Maudio, reasoning, vision
Z AI (Zhipu AI)4No200Kimage, reasoning, text, video
SambaNova4Registration128Kimage, reasoning, text
SiliconFlow3Registration131Ktext
xAI3Registration2Mtext
Chutes.ai2Registration131Kreasoning, text
Glhf.chat2Registration131Ktext
Grok (xAI)2Registration131Ktext
AI21 Labs2Registration256Ktext
DeepSeek2Registration128Ktext
Nscale2Registration128Ktext
Nebius1Registration128Ktext

💰 Renewable Credits

Providers that periodically renew free credits.

ProviderFree ModelsCredit ModelMax ContextModalitiesGet API Key
OpenRouter29Free tier + $10 topup → 1K RPD1Maudio, code, embeddings, image, reasoning, rerank, text, video

Quick Reference — Base URLs & API Keys

ProviderBase URLGet API KeyCredit Card?
NVIDIA NIMhttps://integrate.api.nvidia.com/v1Get Key →Phone verification
Cloudflare Workers AIhttps://api.cloudflare.com/client/v4/accounts/{account_id}/ai/runGet Key →No
ModelScopehttps://api-inference.modelscope.cn/v1Get Key →Registration
OpenRouterhttps://openrouter.ai/api/v1Get Key →Registration
GitHub Modelshttps://models.github.ai/inferenceGet Key →No
OVHcloud AI Endpointshttps://oai.endpoints.kepler.ai.cloud.ovh.net/v1Get Key →Registration
Groqhttps://api.groq.com/openai/v1Get Key →No
Google Geminihttps://generativelanguage.googleapis.com/v1betaGet Key →No
Mistral AIhttps://api.mistral.ai/v1Get Key →No
LLM7.iohttps://api.llm7.io/v1Get Key →No
Cerebrashttps://api.cerebras.ai/v1Get Key →No
Coherehttps://api.cohere.com/v2Get Key →No
Ollama Cloudhttps://api.ollama.comGet Key →Registration
Agnes AIhttps://apihub.agnes-ai.com/v1Get Key →Registration
Aion Labshttps://api.aionlabs.ai/v1Get Key →Registration
Hugging Facehttps://router.huggingface.co/v1Get Key →No
Kilo Codehttps://api.kilo.ai/api/gatewayGet Key →No
Alibaba Cloud Model Studiohttps://dashscope-intl.aliyuncs.com/compatible-mode/v1Get Key →Registration
OpenCode Zenhttps://opencode.ai/zen/v1Get Key →Registration
Z AI (Zhipu AI)https://open.bigmodel.cn/api/paas/v4Get Key →No
SambaNovahttps://api.sambanova.ai/v1Get Key →Registration
SiliconFlowhttps://api.siliconflow.cn/v1Get Key →Registration
xAIhttps://api.x.ai/v1Get Key →Registration
Chutes.aihttps://api.chutes.ai/v1Get Key →Registration
Glhf.chathttps://glhf.chat/api/openai/v1Get Key →Registration
Grok (xAI)https://api.x.ai/v1Get Key →Registration
AI21 Labshttps://api.ai21.com/studio/v1Get Key →Registration
DeepSeekhttps://api.deepseek.com/v1Get Key →Registration
Nscalehttps://inference.api.nscale.com/v1Get Key →Registration
Nebiushttps://api.studio.nebius.com/v1Get Key →Registration

Best Free Models by Provider

ProviderBest Free ModelModel IDMax ContextRate Limit
NVIDIA NIMz-ai/glm-5.2z-ai/glm-5.21MUp to 40 RPM
z-ai/glm-5.1z-ai/glm-5.1202KUp to 40 RPM
baai/bge-m3baai/bge-m3131KUp to 40 RPM
Cloudflare Workers AIMistral 7B@cf/mistral/mistral-7b-instruct-v0.132KSee provider
Qwen 1.5 7B@cf/qwen/qwen1.5-7b-chat32KSee provider
@cf/meta/llama-3.3-70b-instruct-fp8-fast@cf/meta/llama-3.3-70b-instruct-fp8-fast131K10K neurons/day (shared)
ModelScopeMiniMax-M2.5-highspeedMiniMax/MiniMax-M2.5204KSee provider
Qwen/Qwen3.5-35B-A3Bqwen-qwen3-5-35b-a3b131K2,000 RPD total; <=500 ..
Qwen/Qwen3.5-27Bqwen-qwen3-5-27b131K2,000 RPD total; <=500 ..
OpenRouterNVIDIA: Nemotron 3 Ultra (free)nvidia/nemotron-3-ultra-550b-a55b:free1MSee provider
Poolside: Laguna M.1 (free)poolside/laguna-m.1:free262KSee provider
NVIDIA: Nemotron 3 Super (free)nvidia/nemotron-3-super-120b-a12b:free1MSee provider
GitHub ModelsPhi-4Phi-4131KSee provider
Mistral Large (24.11)Mistral-large-2411131KSee provider
AI21 Jamba 1.5 LargeAI21-Jamba-1.5-Large256KSee provider
OVHcloud AI EndpointsQwen3.5-397B-A17Bqwen3.5-397b-a17b131K2 RPM (anonymous)
gpt-oss-20bgpt-oss-20b128K2 RPM (anonymous)
Meta-Llama-3_3-70B-Instructmeta-llama-3_3-70b-instruct131K2 RPM (anonymous)
GroqMoonshot Kimi K2moonshotai/kimi-k2-instruct131KSee provider
Moonshot Kimi K2 0905moonshotai/kimi-k2-instruct-0905131KSee provider
llama-3.3-70b-versatilellama-3.3-70b-versatile131K30 RPM, 1,000 RPD
Google GeminiGemini 3.5 Flashgemini-3.5-flash1M15 RPM, 1,500 RPD
Gemini 3.1 Flash-Litegemini-3.1-flash-lite1M30 RPM, 1,500 RPD
Gemini 2.5 Flashgemini-2.5-flash1M15 RPM, 1,500 RPD
Mistral AIMistral 7Bopen-mistral-7b32KSee provider
Mixtral 8x7Bopen-mixtral-8x7b32KSee provider
Mistral Medium 3.5 (128B)mistral-medium-3-5-128b256K~1 RPS, 500K TPM
LLM7.iodeepseek-r1-0528deepseek-r1-0528131K30 RPM (120 with token)
deepseek-v3-0324deepseek-v3-0324131K30 RPM (120 with token)
gemini-2.5-flash-litegemini-2-5-flash-lite131K30 RPM (120 with token)
CerebrasLlama 3.1 70Bllama3.1-70b131KSee provider
gpt-oss-120bgpt-oss-120b128K30 RPM, 14,400 RPD, 1M ..
zai-glm-4.7zai-glm-4.7128K10 RPM, 100 RPD, 1M TPD
CohereCommand A+ (218B)command-a-218b128K20 RPM
Command A (111B)command-a-111b256K20 RPM
Command R+command-r128K20 RPM
Ollama Cloudgpt-oss:120b-cloudgpt-oss-120b-cloud128KSession/weekly limits (..
deepseek-v3.1:671b-clouddeepseek-v3-1-671b-cloud128KSession/weekly limits (..
qwen3-coder:480b-cloudqwen3-coder-480b-cloud128KSession/weekly limits (..
Agnes AIagnes-1.5-flashagnes-1.5-flash256K30 RPM
agnes-2.0-flashagnes-2.0-flash256K30 RPM
agnes-image-2.0-flashagnes-image-2.0-flash4K30 RPM (1K)
Aion LabsAion 2.5aion-2-5128K15 RPM, 20K TPD
Aion 2.0aion-2-0128K15 RPM, 20K TPD
Aion-RP 1.0 (8B)aion-rp-1-0-8b32K15 RPM, 20K TPD
Hugging FaceMeta-Llama-3.1-8B-Instructmeta-llama-3-1-8b-instruct128KCredit-metered
Mistral-7B-Instruct-v0.3mistral-7b-instruct-v0-332KCredit-metered
Mixtral-8x7B-Instruct-v0.1mixtral-8x7b-instruct-v0-132KCredit-metered
Kilo Codex-ai/grok-code-fast-1:freex-ai-grok-code-fast-1-free256K~200 req/hr
minimax/minimax-m2.5:freeminimax/minimax-m2.5196K~200 req/hr
bytedance-seed/dola-seed-2.0-pro:freebytedance-seed-dola-seed-2-0-pro-free131K~200 req/hr
Alibaba Cloud Model StudioQwen3-Maxqwen3-max128KTiered by region
Qwen3-Plusqwen3-plus1MTiered by region
Qwen3-VL-Plusqwen3-vl-plus128KTiered by region
OpenCode Zenbig-picklebig-pickle0
DeepSeek V4 Flashdeepseek-v4-flash-free1M
MiMo-V2.5mimo-v2.5-free1M
Z AI (Zhipu AI)GLM-4.7-Flashglm-4.7-flash200K1 concurrent request
GLM-4.6V-Flashglm-4-6v-flash128K1 concurrent request
GLM-4.5-Flashglm-4-5-flash128K1 concurrent request
SambaNovaDeepSeek-V3.1deepseek-v3-1128K20 RPM, 20 RPD, 200K TPD
DeepSeek-V3.2 (Preview)deepseek-v3-2-preview128K20 RPM, 20 RPD, 200K TPD
MiniMax-M2.7minimax-m2-7128K20 RPM, 20 RPD, 200K TPD
SiliconFlowdeepseek-ai/DeepSeek-R1-Distill-Qwen-7Bdeepseek-ai-deepseek-r1-distill-qwen-7b131K30 RPM, 60K TPM
Abbreviationabbreviation131KSee provider
deepseek-ai/DeepSeek-OCRdeepseek-ai-deepseek-ocr131K30 RPM, 60K TPM
xAIgrok-4.3grok-4-31MCredit-based
grok-4.1-fastgrok-4-1-fast2MCredit-based
grok-3-minigrok-3-mini131KCredit-based
Chutes.aiDeepSeek-R1deepseek-ai/DeepSeek-R1131KCommunity-powered, no h..
Llama 3.1 70Bmeta-llama/Meta-Llama-3.1-70B-Instruct131KCommunity-powered, no h..
Glhf.chatLlama 3.1 70Bmeta-llama/Meta-Llama-3.1-70B-Instruct131KUnlimited for free models
Mixtral 8x7Bmistralai/Mixtral-8x7B-Instruct-v0.132KUnlimited for free models
Grok (xAI)Grok-2grok-2131K$25/month free credits,..
Grok-2 Minigrok-2-mini131K$25/month free credits,..
AI21 LabsJamba Large 1.7jamba-large-1-7256K200 RPM, 10 RPS
Jamba Mini 2jamba-mini-2256K200 RPM, 10 RPS
DeepSeekdeepseek-chat (V3.2)deepseek-chat-v3-2128KDynamic
deepseek-reasoner (R1)deepseek-reasoner-r1128KDynamic
NscaleLlama-3.3-70B-Instructllama-3-3-70b-instruct128KFair-use
DeepSeek-R1-Distill-Llama-70Bdeepseek-r1-distill-llama-70b128KFair-use
NebiusQwen3-235B-A22Bqwen3-235b-a22b128KTier-based

🖥️ Local / Self-Hosted (Unlimited, Private, Free Forever)

ToolTypeHighlights
OllamaCLI + API100+ models, GPU acceleration, OpenAI-compatible endpoint
LM StudioDesktop GUIAny GGUF model, built-in model browser, offline
llama.cppC/C++ engineRuns any GGUF, minimal dependencies
GPT4AllDesktop appCPU-only, no GPU required, open source
Jan.aiDesktop appPrivacy-focused, 100% offline ChatGPT alternative
KoboldCppSingle executableOptimized for creative writing, GGUF

Top Free Models (by Weekly Usage)

Data from freellm.net, updated daily via API monitoring.

ModelProviderContextWeekly Usage
NVIDIA: Nemotron 3 Ultra (free)OpenRouter1M989B tokens
Poolside: Laguna M.1 (free)OpenRouter262K768B tokens
NVIDIA: Nemotron 3 Super (free)OpenRouter1M440B tokens
z-ai/glm-5.2NVIDIA NIM1M228B tokens
z-ai/glm-5.1NVIDIA NIM202K158B tokens
Cohere: North Mini Code (free)OpenRouter256K147B tokens
Poolside: Laguna XS.2 (free)OpenRouter262K90B tokens
Poolside: Laguna XS 2.1 (free)OpenRouter262K81B tokens
baai/bge-m3NVIDIA NIM131K70B tokens
Tencent: Hy3 (free)OpenRouter262K68B tokens

Repository Structure

awesome-free-llm-apis/
├── README.md              ← Complete provider directory & code examples
├── code-examples/          ← Ready-to-use config snippets
│   ├── claude-code.md
│   ├── cursor.md
│   └── codex.md
└── LICENSE                 ← MIT

For the full structured dataset with 453 models and daily updates, visit freellm.net.


Contributing

We welcome contributions!

  • Add a missing free model — Open an issue or submit a PR
  • Fix inaccurate data — Rate limits change, providers graduate. PRs welcome
  • Add a config snippet — Have a working config for a tool we don't cover? Add it to code-examples/

Criteria for inclusion

A model belongs in this list if:

  1. The provider explicitly offers a free tier (not just a trial credit)
  2. The API is publicly accessible (no waitlist, closed beta, or reverse-engineering)
  3. For trial credits: clearly labeled and minimum $1 credit value

Links

License

MIT © open-free-llm-api


Last updated: 2026-07-07

关于 About

134+ free LLM APIs & AI API keys from 40+ providers. Google Gemini, NVIDIA NIM, Groq, OpenRouter & more. One-click setup for Claude Code, Cursor and Codex.
free-llm-apifreeairesourcesfreellmfreellmapifreellmmodels

语言 Languages

提交活跃度 Commit Activity

代码提交热力图
过去 52 周的开发活跃度
89
Total Commits
峰值: 20次/周
Less
More

核心贡献者 Contributors