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

Free-LLM — Open Directory of Free AI & LLM APIs

210+ free LLM models from 47 providers — find, compare & configure free models in seconds, plus 9 local/self-hosted tools for unlimited private use.

🌐 Live at free-llm.com — Compare providers · Submit a provider · Guides · Hall of Fame

Website License Community Driven

🌐 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 changelogs, signing up for five platforms just to compare rate limits, or guessing which provider still has a free tier this month.

This repo — backed by the live directory at free-llm.com — is a structured, community-maintained reference covering every provider that lets you use LLMs at zero cost.

  • ✅ Community-maintained — votes, submissions, and edit suggestions from real users, moderated before publishing
  • ✅ Credit card transparency — every provider below is labeled with whether it needs a card, phone verification, or nothing at all
  • ✅ Ready-to-run code — Python/JavaScript/curl snippets for all 33 providers in code-examples/, plus per-tool configs for Claude Code, Cursor, Codex, and OpenCode
  • ✅ Side-by-side comparison — free-llm.com/compare puts two providers head-to-head on limits, models, and pricing

How to Use — 3 Steps

  1. Pick a provider — see the Provider Directory below. New to this? Start with Groq (no credit card, 30 RPM / 14,400 requests per day, free forever).
  2. Get your API key — every row links straight to the provider's key page in Quick Reference. Most only need an email address.
  3. Plug it in — copy the base URL + a model ID from the tables below into the snippets in Quick Start.

Full details, live status, and community notes for each provider live on its page at free-llm.com/provider/<slug> (e.g. free-llm.com/provider/groq).


Quick Start — Use Any Free API in 30 Seconds

Most providers below expose an OpenAI-compatible endpoint. Any tool that accepts a baseURL + apiKey works — just swap the two.

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",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
# Groq free tier: 30 RPM, 14,400 requests/day — generous for personal use

Coding assistants

Point your AI coding tool at a free backend instead of a paid one:

Every other provider has a ready-to-copy snippet in code-examples/ — see Code Examples below.


Provider Directory

⚡ Permanent Free Tiers

Ongoing free access with rate-limited quotas that never expire.

ProviderCredit Card?Rate LimitDaily LimitMonthly LimitKey Models
Google AI StudioNo5-30 RPM (varies by model)Varies by model (Flash / Flash-Lite only; Pro models are paid)Free of chargeGemini 3.1 Flash-Lite, Gemini 3.5 Flash-Lite, Gemini 3.6 Flash, Gemini 3.8 Flash
Mistral (La Plateforme)Phone verification1 request/second-FreeCodestral 2508, Ministral 3 8B, Ministral 3 14B, Mistral Large 3
CohereNo20 requests/minute1,000 API calls/month (trial key)1,000 API calls/month (trial key, non-commercial)North Mini Code, Command A Reasoning, Command R+ (08-2024), Command R (08-2024)
NVIDIA NIMPhone verification~40 RPM (varies per model, not published)--Nemotron 3 Ultra 550B A55B, Nemotron 3.5 Lightning 30B A3B, DeepSeek V4 Pro (0813), Kimi K3
GroqNo30 RPM; 8K TPM on most chat models (varies by model)1,000 requests/day and 200K tokens/day on gpt-oss / Qwen (varies by model)Free ForeverQwen3.8 27B, MiniMax M2.7, Whisper Large v3, Whisper Large v3 Turbo
Z.AI (GLM)Registration~1 request/second (Flash models)~1,000 requests/day (Flash tier)Free tier ongoing, subject to changeGLM-4.6V-Flash (vision), GLM-4.5-Flash, GLM-4.7-Flash
CozeRegistrationVaries by modelToken-based daily limitsResets dailyGPT-4o (via Coze), Gemini 1.5 Pro (via Coze)
OVH AI EndpointsRegistration2 RPM per IP per model (anonymous, free); 400 RPM with a key (pay-as-you-go)UnspecifiedBeta AccessLlama 3.3 70B Instruct, GPT OSS 20B, GPT OSS 120B, Qwen3.5 9B
Cloudflare Workers AINoVaries by model10,000 neurons/day~300,000 neurons/monthGLM 4.7 Flash, Mistral Small 3.1 24B, Qwen3 30B A3B, Gemma 4 26B A4B
Nous PortalNoStandard rate limits (figures not published)Free models only (no monthly credit)Free tier: $0/month. A GitHub user reported (14 Sep 2026) that a payment method (Stripe, $0 charge) is required to get a key - verifyPoolside Laguna S 2.1, StepFun Step 3.7 Flash, Upstage Solar Mini 4, Meituan LongCat 2.0
Hetzner Inference APINo10 requests, 4M input / 100K output tokens per 60s (per API key)No daily cap published (free while the API is experimental)Free during experimental phase, no billing system yetQwen3.6 35B A3B
Pollinations.aiNoAPI key required (anonymous access removed); no cardFree usage paid from Quest Pollen earned in the dashboardFree, no billing systemGPT-OSS 20B, Mistral Large 3, Kimi K2.6, GLM 5.3 Flash
ModelScopePhone verification500 requests/day per model2,000 requests/day totalFree, no billingSee provider
Ollama CloudNo1 concurrent request; starter models onlyMonthly starter allowance (resets monthly from sign-up date)Weekly usage limit resets every 7 daysGemma 4, DeepSeek V4.1 Flash, GPT-OSS 120B (Cloud), GPT-OSS 20B (Cloud)
SiliconFlowPhone verificationFixed limits for free models — exact figures require login, verify on cloud.siliconflow.cn/modelsNot fully published — verify on docs.siliconflow.cnFree models available after identity verificationGLM-4 9B 0414, DeepSeek R1 0528 Qwen3 8B, Qwen3.5 4B, GLM 5.3 Flash
Aion LabsNo15 RPM, 20K TPM (free tier)20,000 tokens/day (daily credit allowance, no card)Free, no billingSee provider
Kilo AI GatewayNo200 requests/hour per IP on free models (anonymous use allowed)Default free models onlyFree plan, no expiryinclusionAI Ling 3.1 Flash, Cohere North Mini Code, Thinking Machines Inkling Small, Nemotron 3.5 Lightning
Api.AirforceNo1 RPM1,000 requests/dayFree plan, $0/monthSee provider
RoutewayRegistration5 RPM on :free models200 requests/dayStarter plan free, shared queueMuse Glimmer 30B, MiniMax M2.7, DeepSeek V4 Flash, Step 3.7 Flash
LLM7.ioNo1 req/s, 60/min, 250/hour (free token from dash.llm7.io)100,000 tokens per 24h (free token; may be reduced without notice)Free, no billingNemotron 3 Nano 30B, GLM 5.3 Flash, Mistral Nemo Instruct 2407, Codestral Latest

💰 Renewable Credits

Free access that renews periodically, no one-time expiry.

ProviderCredit Card?Rate LimitFree OfferKey Models
OpenRouterNo20 requests/minute50 requests/day (up to 1000 with $10 topup)Liquid LFM 2.5 2.6B, Nemotron 3 Nano Omni 30B A3B Reasoning, Gemma 4 26B A4B IT, Gemma 4 31B IT
RequestyNoConcurrency-limited (no RPM cap published)200 requests/day (new orgs) / 1,000/day (paying orgs), shared across free modelsMistral Leanstral 1.5, Meta Muse Glimmer 30B, inclusionAI Ling 3.1 Flash, Gemma 4 31B IT
Token HarborRegistrationNot publishedFree allowance, resets on a rolling 7-day periodDeepSeek V4.1 Flash, MiMo V2.6 Flash, TH-Rudder
Vercel AI GatewayRegistrationRate limited per model (lower than paid tier)Monthly free credit (~$5, verify)See provider
OnomeoRegistration12 requests/minute per key30 free-model calls every 5 hours per accountAuto (best free model), DeepSeek V4 Flash, DeepSeek V4 Pro, GLM 5.3 Flash

🎁 One-Time Trial Credits

Sign up and receive credits to use until depleted.

ProviderCredit Card?Credit AmountExpiryKey Models
ReplicateRegistrationLimited free runs on select modelsOne-timeSee provider
Fireworks AIRegistration$1One-timeGLM 5.3, DeepSeek V4.1 Flash, Kimi K3
Nebius (Token Factory)Registration$1 trial credit, valid 30 days (bank card required)One-timeQwen3 235B A22B Instruct 2507, GLM 5.3, GPT OSS 120B, Qwen3.8 27B
CerebrasRegistration$5 after adding a verified payment method (expires after 30 days)30 daysQwen 3.8 27B, GPT OSS 120B
Novita AIRegistrationNo automatic sign-up credit; $10 per referral (both sides)One-timeGLM 5.3, Kimi K3, DeepSeek V4.1 Flash, Qwen3.8 27B
Scaleway Generative APIsRegistration1M tokensOne-timeGemma 4 26B A4B IT, GPT-OSS 120B, Qwen3.6 35B A3B, DeepSeek V4 Flash (0731)
Qwen (Alibaba)Registration1M tokens/modelOne-time per modelQwen3.8 Flash, Qwen3.7 Plus, Qwen3.8 Max, Qwen3.7 Max
AI21 LabsRegistration$10 (valid 7 days, no card)3 monthsJamba Large, Jamba Mini
UpstageRegistration$103 monthsSee provider
CerebriumRegistration$30One-timeSee provider
NscaleNoMinimum $5 top-up to start (no sign-up credit)One-timeSee provider
Friendli AIRegistrationTrial credits at onboarding (amount not published)One-timeDeepSeek V3.2, Gemma 4 31B, GLM 5.2, GLM 5.3 Flash
GrokifiedNo$5One-timegrok-4.6, grok-4.3, grok-build-0.1

🖥️ 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
llamafileSingle executableMulti-platform, combines llama.cpp + Cosmopolitan Libc
Text Generation WebUIGradio UIHighly customizable, advanced local experimentation
BentoMLInference platformDeploy any AI/ML model anywhere, production-grade

Quick Reference — Base URLs & API Keys

ProviderBase URLGet API Key
OpenRouterhttps://openrouter.ai/api/v1Get Key →
Google AI Studiohttps://generativelanguage.googleapis.com/v1betaGet Key →
Mistral (La Plateforme)https://api.mistral.ai/v1Get Key →
Coherehttps://api.cohere.ai/v1Get Key →
Replicatehttps://api.replicate.com/v1Get Key →
Fireworks AIhttps://api.fireworks.ai/inference/v1Get Key →
NVIDIA NIMhttps://integrate.api.nvidia.com/v1Get Key →
Nebius (Token Factory)https://api.tokenfactory.nebius.com/v1Get Key →
Cerebrashttps://api.cerebras.ai/v1Get Key →
Groqhttps://api.groq.com/openai/v1Get Key →
Novita AIhttps://api.novita.ai/v3/openaiGet Key →
Scaleway Generative APIshttps://api.scaleway.ai/v1Get Key →
Qwen (Alibaba)https://dashscope-intl.aliyuncs.com/api/v1Get Key →
AI21 Labshttps://api.ai21.com/studio/v1Get Key →
Upstagehttps://api.upstage.ai/v1/solarGet Key →
Z.AI (GLM)https://api.z.ai/api/paas/v4Get Key →
Cozehttps://api.coze.com/v1Get Key →
OVH AI Endpointshttps://oai.endpoints.kepler.ai.cloud.ovh.net/v1Get Key →
Cloudflare Workers AIhttps://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run/Get Key →
Nous Portalhttps://inference-api.nousresearch.com/v1Get Key →
Hetzner Inference APIhttps://inference.hetzner.com/api/v1Get Key →
Pollinations.aihttps://text.pollinations.aiGet Key →
Requestyhttps://router.requesty.ai/v1Get Key →
ModelScopehttps://api-inference.modelscope.cn/v1Get Key →
Token Harborhttps://tokenharbor.ai/v1Get Key →
Cerebriumhttps://api.cortex.cerebrium.ai/v4Get Key →
Ollama Cloudhttps://ollama.com/v1Get Key →
SiliconFlowhttps://api.siliconflow.com/v1Get Key →
Aion Labshttps://api.aionlabs.ai/v1Get Key →
Nscalehttps://inference.api.nscale.com/v1Get Key →
Kilo AI Gatewayhttps://api.kilo.ai/api/gatewayGet Key →
Vercel AI Gatewayhttps://ai-gateway.vercel.sh/v1Get Key →
Api.Airforcehttps://api.airforce/v1Get Key →
Routewayhttps://api.routeway.ai/v1Get Key →
Onomeohttps://onomeo.com/v1Get Key →
Friendli AIhttps://inference.friendli.ai/v1Get Key →
Grokifiedhttps://api.grokified.com/v1Get Key →
LLM7.iohttps://api.llm7.io/v1Get Key →

Guides & Tutorials

Published at free-llm.com/guides:

  • Best Free LLM APIs in 2026 — side-by-side comparison of top picks
  • Gemini vs ChatGPT (Free Tier) — what you actually get for $0
  • How to Use OpenRouter — setup walkthrough with code
  • OpenRouter Alternatives — other aggregators worth trying
  • Local LLMs with Ollama — get started in under 5 minutes
  • Ultimate Free LLM API Guide — the comprehensive deep-dive

Community Features

Free-LLM is community-driven. The website at free-llm.com lets visitors:

  • Vote on providers to surface the most useful ones
  • Submit new providers and models
  • Propose edits to existing provider data (admin-reviewed)
  • Report models that have gone from free to paid
  • Earn recognition on the Hall of Fame leaderboard

Data syncs back to this repository.


Code Examples

The code-examples/ directory has ready-to-run Python, JavaScript, and curl snippets — just add your API key.

By coding assistant: Claude Code · Cursor · Codex CLI · OpenCode

By provider (31): AI21 Labs · Aion Labs · Cerebras · Cerebrium · Cloudflare Workers AI · Cohere · Coze · Fireworks AI · Friendli AI · Google AI Studio · Groq · Hetzner Inference API · LLM7.io · Mistral (La Plateforme) · ModelScope · Nebius (Token Factory) · Nous Portal · Novita AI · Nscale · NVIDIA NIM · Ollama Cloud · OpenRouter · OVH AI Endpoints · Pollinations.ai · Qwen (Alibaba) · Replicate · Requesty · Scaleway Generative APIs · SiliconFlow · Upstage · Z.AI (GLM)

Local / Self-Hosted: BentoML · GPT4All · Jan.ai · KoboldCpp · llama.cpp · llamafile · LM Studio · Ollama · Text Gen WebUI


Repository Structure

Free-LLM/
├── README.md                 ← You are here (English)
├── README.zh-CN.md            ← 简体中文
├── README.zh-TW.md            ← 繁體中文
├── README.ja.md               ← 日本語
├── README.ko.md                ← 한국어
├── CONTRIBUTING.md            ← Contribution guidelines
├── code-examples/             ← Ready-to-use snippets (per-provider + per-tool)
├── .github/                   ← Issue/PR templates
└── LICENSE                    ← MIT

Contributing

See CONTRIBUTING.md for the full guide. Quick version:

  1. Add a provider — use the submit form on the website, or open an issue/PR here.
  2. Fix inaccurate data — rate limits change, providers graduate or shut down. PRs welcome.
  3. Add a config snippet — have a working config for a tool we don't cover? Add it to code-examples/.
  4. Vote & discuss — help the community surface the best options at free-llm.com.

Criteria for inclusion

A provider belongs in this list if:

  1. It explicitly offers a free tier (not just a trial credit with no free-forever option) — see Provider Directory for how we split permanent tiers from one-time credits
  2. The API is publicly accessible (no waitlist, closed beta, or reverse-engineering)
  3. For trial credits: clearly labeled and the free-forever alternative (if any) is called out

Links

Star History

Star History Chart


License

MIT — see LICENSE for details.


Data synced automatically from the live directory — last updated: 2026-10-15

关于 About

Directory of 34+ free LLM & AI APIs — permanent free tiers, trial credits, and no-card options. GPT-4o, Gemini, Claude, Llama, DeepSeek, Mistral & more. Synced daily from the live directory at free-llm.com.
chatgpt-alternativeclaude-4-6-opusclaude-apideepseek-freefree-apisfree-llmsgemini-apillama-apillm-apisopenai-alternative

语言 Languages

JavaScript34.4%
Python33.9%
Shell29.8%
PHP1.9%

提交活跃度 Commit Activity

代码提交热力图
过去 52 周的开发活跃度
109
Total Commits
峰值: 14次/周
Less
More

核心贡献者 Contributors