🌟 Free LLM Hub

🌍 Read this in other languages: 🇺🇸 English • 🇧🇷 Português (BR) • 🇵🇹 Português (PT) • 🇪🇸 Español • 🇫🇷 Français • 🇨🇳 中文 (简体) • 🇯🇵 日本語

A unified, community-driven catalog of LLM APIs, inference engines, gateways, and the entire OSS LLM ecosystem.

Total entries: 187 • Last updated: auto-generated

📑 Table of Contents

• 📡 Provider APIs (18)
• 🔌 Inference Providers (24)
• 💰 Subscription Plans (47)
• 🛠️ Inference Engines (OSS) (13)
• 🚪 Gateways / Routers (11)
• 🎨 Specialty APIs (22)
• 🤖 Agent Frameworks (5)
• 📚 LLM Frameworks (6)
• 🗄️ Vector Databases (8)
• 📊 Eval Frameworks (4)
• 📦 Model Catalogs (4)
• 💻 Coding Tools (5)
• 🖥️ Desktop UIs (6)
• 🧬 Open-Weights Families (14)

---

📡 Provider APIs

Name	Country	Pricing	Rate Limits	Models / Notes	Link
Google AI Studio (Gemini)	US	🟢 Freemium	15 RPM, 1000 RPD	gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite	🔗
Mistral AI (La Plateforme)	FR	🟢 Freemium	60 RPM, 1,000,000,000 tok/mo	mistral-large-3, mistral-small-3.1, ministral-8b	🔗
Cohere	CA	🟢 Freemium	20 RPM	command-a, command-r-plus, command-r7b	🔗
Zhipu AI (Z.AI / GLM)	CN	🟢 Freemium	—	glm-4.7-flash, glm-4.5-flash, glm-4.6v-flash	🔗
DeepSeek Platform	CN	💵 $0.14/$0.28 per MTok	—	deepseek-v3.2, deepseek-r1, deepseek-v4-pro	🔗
Moonshot AI (Kimi)	CN	🟢 Freemium	—	kimi-k2.5, kimi-k2.6, kimi-long-context	🔗
DashScope (Alibaba)	CN	🟢 Freemium	—	qwen-max, qwen-plus, qwen-vl	🔗
MiniMax	CN	🎁 Trial	—	minimax-m2.5, minimax-m2.1, abab6.5	🔗
01.AI (Yi / 零一万物)	CN	🎁 Trial	—	yi-large, yi-lightning, yi-vision	🔗
StepFun (阶跃星辰)	CN	🎁 Trial	—	step-3.5-flash, step-2	🔗
Baidu Qianfan (ERNIE)	CN	🟢 Freemium	—	ernie-4.0, ernie-speed	🔗
Tencent Hunyuan	CN	🟢 Freemium	—	hunyuan-lite, hunyuan-pro, hunyuan-turbo	🔗
InternLM (Shanghai AI Lab)	CN	🟢 Freemium	—	internlm2.5, internvl	🔗
OpenAI API	US	💵 $1.25/$10.0 per MTok	—	gpt-5, gpt-5.1, gpt-5.2	🔗
Anthropic API	US	💵 $3.0/$15.0 per MTok	—	claude-sonnet-4.6, claude-opus-4.6, claude-haiku-4	🔗
xAI Grok API	US	💵 $3.0/$15.0 per MTok	—	grok-3, grok-2	🔗
Perplexity Sonar API	US	💵 Pay-per-token	—	sonar, sonar-pro	🔗
Reka AI	US	🎁 Trial	—	reka-core, reka-flash, reka-edge	🔗

🔌 Inference Providers

Name	Country	Pricing	Rate Limits	Models / Notes	Link
Groq	US	🟢 Freemium	30 RPM, 1000 RPD	llama-3.3-70b, llama-4-scout, kimi-k2	🔗
Cerebras	US	🟢 Freemium	30 RPM, 14400 RPD	llama-3.3-70b, qwen3-235b, gpt-oss-120b	🔗
NVIDIA NIM	US	🟢 Freemium	40 RPM	llama-3.3-70b, mistral-large, qwen3-235b	🔗
Cloudflare Workers AI	US	🟢 Freemium	—	llama-3.3-70b, qwen-qwq-32b, +47 more	🔗
HuggingFace Inference Providers	US	🟢 Freemium	—	llama-3.3-70b, qwen2.5-72b, mistral-7b	🔗
OpenCode Zen	US	🟢 Freemium	—	—	🔗
Ollama Cloud	US	🟢 Freemium	—	deepseek-v3.2, qwen3.5, kimi-k2.5	🔗
LLM7.io	GB	🟢 Free	30 RPM	deepseek-r1, qwen2.5-coder, +27 more	🔗
Kluster AI	US	🟢 Freemium	—	deepseek-r1, llama-4-maverick, qwen3-235b	🔗
Together AI	US	🎁 $5 trial	—	llama-3.3, mixtral, qwen-2.5	🔗
Fireworks AI	US	🎁 $1 trial	600 RPM	llama-3.3-70b, qwen-2.5-72b, deepseek-v3	🔗
DeepInfra	US	💵 $0.14/$0.28 per MTok	—	deepseek-v4-flash, kimi-k2.6, glm-5	🔗
Baseten	US	🎁 $30 trial	—	—	🔗
Nebius	NL	🎁 Trial	—	—	🔗
Novita AI	SG	🎁 Trial	—	—	🔗
Hyperbolic	US	🎁 Trial	—	llama-3.3, deepseek	🔗
SambaNova Cloud	US	🟢 Freemium	—	llama-4	🔗
Scaleway Generative APIs	FR	🟢 Freemium	—	—	🔗
Lepton AI	US	🎁 $10 trial	—	—	🔗
Avian.io	US	🟢 Freemium	—	llama-3.1-405b, qwen	🔗
Featherless AI	US	💳 $10/mo	—	4000+ HF models	🔗
Targon (Bittensor)	US	🟢 Freemium	—	deepseek, llama	🔗
Chutes	—	🎁 Trial	—	—	🔗
SiliconFlow (硅基流动)	CN	🟢 Freemium	1000 RPM, 50000 TPM	qwen3-8b, deepseek-r1-distill, glm-4.1v-9b	🔗

💰 Subscription Plans

Name	Country	Pricing	Rate Limits	Models / Notes	Link
ElevenLabs Starter	—	💳 $5/mo	—	30K caracteres TTS/mês	🔗
Suno Pro	—	💳 $10/mo	—	500 créditos diários	🔗
Midjourney Basic	—	💳 $10/mo	—	~200 imagens/mês	🔗
GitHub Copilot Pro	—	💳 $10/mo	—	gpt-5, claude-opus-4.6, gemini-3	🔗
Tabnine Pro	—	💳 $12/mo	—	Code completion full-length, multi-LLM chat	🔗
Leonardo.ai Apprentice	—	💳 $12/mo	—	—	🔗
Descript Hobbyist	—	💳 $12/mo	—	10h transcrição/mês	🔗
Runway Standard	—	💳 $15/mo	—	—	🔗
Windsurf Pro	—	💳 $15/mo	—	claude-opus-4.6, gpt-5.4, gemini-3-pro	🔗
Mistral Le Chat Pro	—	💳 $15/mo	—	mistral-large-3	🔗
Augment Code Indie	—	💳 $15/mo	—	—	🔗
Writesonic	—	💳 $16/mo	—	—	🔗
GitHub Copilot Business	—	💳 $19/mo	—	—	🔗
Amazon Q Developer Pro	—	💳 $19/mo	—	—	🔗
NotebookLM Plus	—	💳 $19.99/mo	—	—	🔗
ChatGPT Plus	—	💳 $20/mo	—	gpt-5.4, codex, dall-e-3	🔗
Claude Pro	—	💳 $20/mo	—	claude-opus-4.6, claude-sonnet-4.6	🔗
Gemini Advanced	—	💳 $20/mo	—	gemini-3-pro	🔗
Perplexity Pro	—	💳 $20/mo	—	—	🔗
Cursor Pro	—	💳 $20/mo	—	claude-opus-4.6, gpt-5.4, gemini-3-pro	🔗
v0 Premium (Vercel)	—	💳 $20/mo	—	—	🔗
Lovable	—	💳 $20/mo	—	—	🔗
Claude Code Pro	—	💳 $20/mo	—	—	🔗
Grok Premium (X)	—	💳 $22/mo	—	grok-3	🔗
HeyGen Creator	—	💳 $29/mo	—	—	🔗
Synthesia Starter	—	💳 $29/mo	—	—	🔗
Midjourney Standard	—	💳 $30/mo	—	—	🔗
Tabnine Enterprise	—	💳 $39/mo	—	Self-host VPC/on-prem	🔗
GitHub Copilot Enterprise	—	💳 $39/mo	—	—	🔗
Cursor Teams	—	💳 $40/mo	—	—	🔗
Jasper Creator	—	💳 $49/mo	—	—	🔗
Copy.ai Pro	—	💳 $49/mo	—	—	🔗
Cursor Pro+	—	💳 $60/mo	—	3x usage Claude/GPT/Gemini	🔗
Windsurf Team	—	💳 $100/mo	—	1500 credits/user, SSO	🔗
ChatGPT Pro (new)	—	💳 $100/mo	—	5x Plus, 10x Codex	🔗
Claude Max 5x	—	💳 $100/mo	—	—	🔗
ChatGPT Pro (original)	—	💳 $200/mo	—	20x Plus, Sora, exclusive Pro models	🔗
Claude Max 20x	—	💳 $200/mo	—	—	🔗
Cursor Ultra	—	💳 $200/mo	—	—	🔗
Windsurf Max	—	💳 $200/mo	—	Unlimited credits, 1M context	🔗
Gemini Ultra	—	💳 $250/mo	—	gemini-3-pro-deep-think	🔗
Devin (Cognition AI)	—	💳 $500/mo	—	Autonomous coding agent	🔗
OpenAI Enterprise	—	💳 Custom	—	Custom pricing, SOC2	🔗
Anthropic Enterprise	—	💳 Custom	—	—	🔗
AWS Bedrock	—	💵 Pay-per-token	—	claude, llama, mistral	🔗
Azure OpenAI	—	💵 Pay-per-token	—	—	🔗
Google Vertex AI	—	💵 Pay-per-token	—	—	🔗

🛠️ Inference Engines (OSS)

Name	Country	Pricing	Rate Limits	Models / Notes	Link
vLLM	—	🏠 Self-hosted	—	—	🔗
Ollama	—	🏠 Self-hosted	—	—	🔗
llama.cpp	—	🏠 Self-hosted	—	—	🔗
Text Generation Inference (TGI)	—	🏠 Self-hosted	—	—	🔗
SGLang	—	🏠 Self-hosted	—	—	🔗
TensorRT-LLM	—	🏠 Self-hosted	—	—	🔗
LocalAI	—	🏠 Self-hosted	—	—	🔗
LMDeploy	—	🏠 Self-hosted	—	—	🔗
MLC-LLM	—	🏠 Self-hosted	—	—	🔗
KTransformers	—	🏠 Self-hosted	—	—	🔗
ExLlamaV2	—	🏠 Self-hosted	—	—	🔗
Aphrodite Engine	—	🏠 Self-hosted	—	—	🔗
CTranslate2	—	🏠 Self-hosted	—	—	🔗

🚪 Gateways / Routers

Name	Country	Pricing	Rate Limits	Models / Notes	Link
OpenRouter	US	🟢 Freemium	20 RPM, 50 RPD	deepseek-r1-free, llama-3.3-70b-free, gpt-oss-120b-free	🔗
GitHub Models	US	🟢 Freemium	15 RPM, 150 RPD	gpt-5, claude-sonnet-4, llama-3.3-70b	🔗
Vercel AI Gateway	US	🟢 Freemium	—	multiple	🔗
LiteLLM	—	🏠 Self-hosted	—	—	🔗
Portkey AI Gateway	—	🟢 Freemium	—	—	🔗
OneAPI	—	🏠 Self-hosted	—	—	🔗
NewAPI	—	🏠 Self-hosted	—	—	🔗
Helicone	—	🟢 Freemium	—	—	🔗
Langfuse	—	🟢 Freemium	—	—	🔗
RouteLLM	—	🏠 Self-hosted	—	—	🔗
Arize Phoenix	—	🏠 Self-hosted	—	—	🔗

🎨 Specialty APIs

Name	Country	Pricing	Rate Limits	Models / Notes	Link
ElevenLabs	US	🟢 Freemium	—	10K chars/mês free	🔗
PlayHT	—	🟢 Freemium	—	12.5K chars/mês free	🔗
Cartesia (Sonic)	—	🟢 Freemium	—	10K chars/mês free	🔗
Resemble AI	—	🎁 Trial	—	—	🔗
Coqui XTTS	—	🏠 Self-hosted	—	—	🔗
Kokoro TTS	—	🏠 Self-hosted	—	—	🔗
Deepgram	—	🎁 $200 trial	—	—	🔗
AssemblyAI	—	🎁 $50 trial	—	—	🔗
Whisper	—	🏠 Self-hosted	—	—	🔗
faster-whisper	—	🏠 Self-hosted	—	—	🔗
Voyage AI	—	🟢 Freemium	—	50M tokens free	🔗
Jina AI	—	🟢 Freemium	—	1M tokens free	🔗
Mixedbread	DE	🟢 Freemium	—	—	🔗
Nomic Atlas	—	🟢 Freemium	—	—	🔗
fal.ai	—	🎁 Trial	—	—	🔗
Pollinations	—	🟢 Free	—	—	🔗
Replicate	—	🎁 $0.5 trial	—	—	🔗
ComfyUI	—	🏠 Self-hosted	—	—	🔗
AUTOMATIC1111 WebUI	—	🏠 Self-hosted	—	—	🔗
Runway	—	💳 $15/mo	—	—	🔗
Kling	CN	🟢 Freemium	—	—	🔗
Black Forest Labs (FLUX)	DE	🎁 Trial	—	—	🔗

🤖 Agent Frameworks

Name	Country	Pricing	Rate Limits	Models / Notes	Link
CrewAI	—	🏠 Self-hosted	—	—	🔗
AutoGen	—	🏠 Self-hosted	—	—	🔗
LangGraph	—	🏠 Self-hosted	—	—	🔗
Pydantic AI	—	🏠 Self-hosted	—	—	🔗
Mastra	—	🏠 Self-hosted	—	—	🔗

📚 LLM Frameworks

Name	Country	Pricing	Rate Limits	Models / Notes	Link
LangChain	—	🏠 Self-hosted	—	—	🔗
LlamaIndex	—	🏠 Self-hosted	—	—	🔗
Haystack	—	🏠 Self-hosted	—	—	🔗
DSPy	—	🏠 Self-hosted	—	—	🔗
Semantic Kernel	—	🏠 Self-hosted	—	—	🔗
Vercel AI SDK	—	🟢 Free	—	—	🔗

🗄️ Vector Databases

Name	Country	Pricing	Rate Limits	Models / Notes	Link
Qdrant	—	🟢 Freemium	—	1GB free cloud	🔗
Weaviate	—	🟢 Freemium	—	—	🔗
Milvus	—	🏠 Self-hosted	—	—	🔗
Chroma	—	🏠 Self-hosted	—	—	🔗
pgvector	—	🏠 Self-hosted	—	—	🔗
Pinecone	—	🟢 Freemium	—	1 free index	🔗
LanceDB	—	🏠 Self-hosted	—	—	🔗
Vespa	—	🏠 Self-hosted	—	—	🔗

📊 Eval Frameworks

Name	Country	Pricing	Rate Limits	Models / Notes	Link
Promptfoo	—	🏠 Self-hosted	—	—	🔗
DeepEval	—	🏠 Self-hosted	—	—	🔗
Ragas	—	🏠 Self-hosted	—	—	🔗
OpenAI Evals	—	🏠 Self-hosted	—	—	🔗

📦 Model Catalogs

Name	Country	Pricing	Rate Limits	Models / Notes	Link
HuggingFace Hub	—	🟢 Freemium	—	—	🔗
ModelScope (Alibaba)	CN	🟢 Freemium	—	—	🔗
models.dev	—	🟢 Free	—	—	🔗
Civitai	—	🟢 Freemium	—	—	🔗

💻 Coding Tools

Name	Country	Pricing	Rate Limits	Models / Notes	Link
Aider	—	🏠 Self-hosted	—	—	🔗
Cline	—	🏠 Self-hosted	—	—	🔗
OpenHands	—	🏠 Self-hosted	—	—	🔗
Continue.dev	—	🏠 Self-hosted	—	—	🔗
Codex CLI	—	🏠 Self-hosted	—	—	🔗

🖥️ Desktop UIs

Name	Country	Pricing	Rate Limits	Models / Notes	Link
Open WebUI	—	🏠 Self-hosted	—	—	🔗
Text Generation WebUI	—	🏠 Self-hosted	—	—	🔗
Jan	—	🏠 Self-hosted	—	—	🔗
GPT4All	—	🏠 Self-hosted	—	—	🔗
LM Studio	—	🟢 Free	—	—	🔗
KoboldCpp	—	🏠 Self-hosted	—	—	🔗

🧬 Open-Weights Families

Name	Country	Pricing	Rate Limits	Models / Notes	Link
Llama (Meta)	—	—	—	llama-3.3-70b, llama-4-scout, llama-4-maverick	🔗
Qwen (Alibaba)	—	—	—	qwen3-0.6b, qwen3-8b, qwen3-72b	🔗
DeepSeek	—	—	—	deepseek-v3, deepseek-r1, deepseek-v4-pro	🔗
Mistral / Mixtral	—	—	—	mistral-7b, mixtral-8x7b, mixtral-8x22b	🔗
Gemma (Google)	—	—	—	gemma-3-1b, gemma-3-4b, gemma-3-12b	🔗
Phi (Microsoft)	—	—	—	phi-4, phi-4-mini	🔗
Yi (01.AI)	—	—	—	yi-6b, yi-9b, yi-34b	🔗
InternLM	—	—	—	internlm2.5-7b, internlm3-20b	🔗
GLM (Zhipu)	—	—	—	glm-4-9b, glm-4-32b, glm-4.5-flash	🔗
Hermes (Nous Research)	—	—	—	hermes-3-405b, hermes-4	🔗
gpt-oss (OpenAI)	—	—	—	gpt-oss-20b, gpt-oss-120b	🔗
Granite (IBM)	—	—	—	granite-3.0, granite-code	🔗
OLMo (AllenAI)	—	—	—	olmo-2-1b, olmo-2-13b, olmo-2-32b	🔗
SmolLM (HuggingFace)	—	—	—	smollm-135m, smollm-360m, smollm-1.7b	🔗

---

🔍 Auto-discovered Models

Auto-generated by scripts/discover_models.py probing public /v1/models endpoints. 9 providers responding publicly, 1157 models total.

🟢 Public endpoints (no auth required)

Provider	Models	Endpoint
openrouter	358	https://openrouter.ai/api/v1
vercel-ai-gateway	274	https://ai-gateway.vercel.sh/v1
deepinfra	141	https://api.deepinfra.com/v1/openai
huggingface-inference	131	https://router.huggingface.co/v1
nvidia-nim	119	https://integrate.api.nvidia.com/v1
novita	104	https://api.novita.ai/v3/openai
kluster	15	https://api.kluster.ai/v1
llm7	8	https://api.llm7.io/v1
sambanova	7	https://api.sambanova.ai/v1

🔒 15 providers require authentication (endpoint valid, key needed)

Provider	Status
cerebras	auth_required_403
dashscope	auth_required_401
deepseek	auth_required_401
fireworks	auth_required_401
github-models	auth_required_401
groq	auth_required_401
minimax	auth_required_401
mistral	auth_required_401
moonshot	auth_required_401
openai-api	auth_required_401
scaleway	auth_required_401
siliconflow	auth_required_401
together	auth_required_401
xai-grok	auth_required_401
zhipu	auth_required_401

⚠️ 3 endpoints with errors (TODO: investigate)

Provider	Status
ollama-cloud	not_found_404
opencode-zen	not_found_404
perplexity-api	not_found_404

---

🤝 Contributing

Edit data/0X-*.yaml, run ./scripts/merge.sh && python scripts/render_readme.py, open a PR.

📜 License

MIT