🌟 Free LLM Hub
A unified, community-driven catalog of LLM APIs, inference engines, gateways, and the entire OSS LLM ecosystem.
Total entries: 188 • Last updated: auto-generated
📑 Table of Contents
📡 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 | — | — | 🔗 |
| Kilo Code Gateway | US | 🟢 Freemium | — | anthropic/claude-opus-4.7, anthropic/claude-sonnet-4.6, o... | 🔗 |
🎨 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 | — | — | 🔗 |
| 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. 10 providers responding publicly, 1528 models total.
🟢 Public endpoints (no auth required)
| Provider | Models | Endpoint |
openrouter | 367 | https://openrouter.ai/api/v1 |
kilo-gateway | 356 | https://api.kilo.ai/api/gateway |
vercel-ai-gateway | 276 | https://ai-gateway.vercel.sh/v1 |
deepinfra | 151 | https://api.deepinfra.com/v1/openai |
nvidia-nim | 131 | https://integrate.api.nvidia.com/v1 |
huggingface-inference | 117 | https://router.huggingface.co/v1 |
novita | 101 | https://api.novita.ai/v3/openai |
kluster | 15 | https://api.kluster.ai/v1 |
sambanova | 9 | https://api.sambanova.ai/v1 |
llm7 | 5 | https://api.llm7.io/v1 |
🔒 14 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 | | `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 | ⚠️ 4 endpoints with errors (TODO: investigate)
| Provider | Status | |----------|--------| | `minimax` | empty_response (200 but no models) | | `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