Supported Providers
Liter-llm supports 165 providers out of the box. Route requests to any provider using the provider/model prefix convention – for example, openai/gpt-4o routes to OpenAI and anthropic/claude-3-opus routes to Anthropic. No extra configuration is needed beyond setting the provider’s API key.
| Provider | Prefix | Chat | Embeddings | Image | Audio | Moderation |
|---|---|---|---|---|---|---|
| A2A | a2a/ |
:white_check_mark: | – | – | – | – |
| Abliteration | abliteration/ |
:white_check_mark: | – | – | – | – |
| AI/ML API | aiml/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | – | – |
| AI21 | ai21/ |
:white_check_mark: | – | – | – | – |
| AI21 Chat | ai21_chat/ |
:white_check_mark: | – | – | – | – |
| AIHubMix | aihubmix/ |
:white_check_mark: | – | – | – | – |
| Amazon Nova | amazon_nova/ |
:white_check_mark: | – | – | – | – |
| Anthropic | anthropic/ |
:white_check_mark: | – | – | – | – |
| Anthropic Text | anthropic_text/ |
:white_check_mark: | – | – | – | – |
| Apertis | apertis/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| AssemblyAI | assemblyai/ |
:white_check_mark: | – | – | :white_check_mark: | – |
| Auto Router | auto_router/ |
:white_check_mark: | – | – | – | – |
| AWS - Bedrock | bedrock/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| AWS - Bedrock Mantle | bedrock_mantle/ |
:white_check_mark: | – | – | – | – |
| AWS - Polly | aws_polly/ |
– | – | – | :white_check_mark: | – |
| AWS - Sagemaker | sagemaker/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| AWS S3 Vectors | s3_vectors/ |
– | – | – | – | – |
| Azure | azure/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| Azure AI | azure_ai/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| Azure AI Document Intelligence | azure_ai/doc-intelligence/ |
– | – | – | – | – |
| Azure AI Foundry Agents | azure_ai/agents/ |
:white_check_mark: | – | – | – | – |
| Azure Text | azure_text/ |
:white_check_mark: | – | – | :white_check_mark: | :white_check_mark: |
| Baseten | baseten/ |
:white_check_mark: | – | – | – | – |
| Black Forest Labs | black_forest_labs/ |
– | – | :white_check_mark: | – | – |
| Brave Search | brave/ |
– | – | – | – | – |
| Bytez | bytez/ |
:white_check_mark: | – | – | – | – |
| Cerebras | cerebras/ |
:white_check_mark: | – | – | – | – |
| Charity Engine | charity_engine/ |
:white_check_mark: | – | – | – | – |
| ChatGPT Subscription | chatgpt/ |
:white_check_mark: | – | – | – | – |
| Chutes | chutes/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Clarifai | clarifai/ |
:white_check_mark: | – | – | – | – |
| Cloudflare AI Workers | cloudflare/ |
:white_check_mark: | – | – | – | – |
| Codestral | codestral/ |
:white_check_mark: | – | – | – | – |
| Cohere | cohere/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Cohere Chat | cohere_chat/ |
:white_check_mark: | – | – | – | – |
| CometAPI | cometapi/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | – | – |
| CompactifAI | compactifai/ |
:white_check_mark: | – | – | – | – |
| Crusoe | crusoe/ |
:white_check_mark: | – | – | – | – |
| Cursor BYOK | cursor/ |
:white_check_mark: | – | – | – | – |
| Custom | custom/ |
:white_check_mark: | – | – | – | – |
| Custom OpenAI | custom_openai/ |
:white_check_mark: | – | – | :white_check_mark: | :white_check_mark: |
| Darkbloom | darkbloom/ |
:white_check_mark: | – | – | – | – |
| Dashscope | dashscope/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | – | – |
| Databricks | databricks/ |
:white_check_mark: | – | – | – | – |
| DataForSEO | dataforseo/ |
– | – | – | – | – |
| DataRobot | datarobot/ |
:white_check_mark: | – | – | – | – |
| Deepgram | deepgram/ |
:white_check_mark: | – | – | :white_check_mark: | – |
| DeepInfra | deepinfra/ |
:white_check_mark: | – | – | – | – |
| Deepseek | deepseek/ |
:white_check_mark: | – | – | – | – |
| Docker Model Runner | docker_model_runner/ |
:white_check_mark: | – | – | – | – |
| DuckDuckGo | duckduckgo/ |
– | – | – | – | – |
| ElevenLabs | elevenlabs/ |
:white_check_mark: | – | – | :white_check_mark: | – |
| Empiriolabs | empiriolabs/ |
:white_check_mark: | – | – | – | – |
| Empower | empower/ |
:white_check_mark: | – | – | – | – |
| Exa AI | exa_ai/ |
– | – | – | – | – |
| Fal AI | fal_ai/ |
:white_check_mark: | – | :white_check_mark: | – | – |
| Featherless AI | featherless_ai/ |
:white_check_mark: | – | – | – | – |
| Firecrawl | firecrawl/ |
– | – | – | – | – |
| Fireworks AI | fireworks_ai/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| FriendliAI | friendliai/ |
:white_check_mark: | – | – | – | – |
| Galadriel | galadriel/ |
:white_check_mark: | – | – | – | – |
| GigaChat | gigachat/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| GitHub Copilot | github_copilot/ |
:white_check_mark: | – | – | – | – |
| GitHub Models | github/ |
:white_check_mark: | – | – | – | – |
| GMI Cloud | gmi/ |
:white_check_mark: | – | – | – | – |
| Google - Distributed Cloud (Gemini) | gdc/ |
:white_check_mark: | – | – | – | – |
| Google - Vertex AI | vertex_ai/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | – |
| Google AI Studio - Gemini | gemini/ |
:white_check_mark: | – | – | – | – |
| Google PSE | google_pse/ |
– | – | – | – | – |
| GradientAI | gradient_ai/ |
:white_check_mark: | – | – | – | – |
| Groq AI | groq/ |
:white_check_mark: | – | – | :white_check_mark: | – |
| Helicone | helicone/ |
:white_check_mark: | – | – | – | – |
| Heroku | heroku/ |
:white_check_mark: | – | – | – | – |
| Hosted VLLM | hosted_vllm/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Huggingface | huggingface/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Hyperbolic | hyperbolic/ |
:white_check_mark: | – | – | – | – |
| IBM - Watsonx.ai | watsonx/ |
:white_check_mark: | :white_check_mark: | – | :white_check_mark: | – |
| Inception Labs | inception/ |
:white_check_mark: | – | – | – | – |
| Infinity | infinity/ |
– | :white_check_mark: | – | – | – |
| Jina AI | jina_ai/ |
– | :white_check_mark: | – | – | – |
| Lambda AI | lambda_ai/ |
:white_check_mark: | – | – | – | – |
| LangGraph | langgraph/ |
:white_check_mark: | – | – | – | – |
| Lemonade | lemonade/ |
:white_check_mark: | – | – | – | – |
| LibertAI | libertai/ |
:white_check_mark: | – | – | – | – |
| Linkup | linkup/ |
– | – | – | – | – |
| LiteLLM Proxy | litellm_proxy/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | – | – |
| llama.cpp | llamacpp/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Llamafile | llamafile/ |
:white_check_mark: | – | – | – | – |
| LlamaGate | llamagate/ |
:white_check_mark: | – | – | – | – |
| LM Studio | lmstudio/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| LocalAI | localai/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | – |
| Manus | manus/ |
:white_check_mark: | – | – | – | – |
| Maritalk | maritalk/ |
:white_check_mark: | – | – | – | – |
| Meta - Llama API | meta_llama/ |
:white_check_mark: | – | – | – | – |
| Meta - Model API | meta/ |
:white_check_mark: | – | – | – | – |
| Milvus | milvus/ |
– | – | – | – | – |
| Minimax | minimax/ |
:white_check_mark: | – | – | – | – |
| Mistral AI API | mistral/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| ModelScope | modelscope/ |
:white_check_mark: | – | :white_check_mark: | – | – |
| Moonshot | moonshot/ |
:white_check_mark: | – | – | – | – |
| Morph | morph/ |
:white_check_mark: | – | – | – | – |
| NanoGPT | nanogpt/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Nebius AI Studio | nebius/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Neosantara | neosantara/ |
:white_check_mark: | – | – | – | – |
| NLP Cloud | nlp_cloud/ |
:white_check_mark: | – | – | – | – |
| Novita AI | novita/ |
:white_check_mark: | – | – | – | – |
| Nscale | nscale/ |
:white_check_mark: | – | – | – | – |
| Nvidia NIM | nvidia_nim/ |
:white_check_mark: | – | – | – | – |
| Nvidia Riva | nvidia_riva/ |
– | – | – | :white_check_mark: | – |
| OCI | oci/ |
:white_check_mark: | – | – | – | – |
| Ollama | ollama/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Oobabooga | oobabooga/ |
:white_check_mark: | – | – | :white_check_mark: | :white_check_mark: |
| OpenAI | openai/ |
:white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| OpenAI-like | openai_like/ |
– | :white_check_mark: | – | – | – |
| OpenCode Go | opencode-go/ |
:white_check_mark: | – | – | – | – |
| OpenCode Zen | opencode/ |
:white_check_mark: | – | – | – | – |
| OpenRouter | openrouter/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| OVHCloud AI Endpoints | ovhcloud/ |
:white_check_mark: | :white_check_mark: | – | :white_check_mark: | – |
| Parallel AI | parallel_ai/ |
– | – | – | – | – |
| Parasail | parasail/ |
:white_check_mark: | – | – | – | – |
| Perplexity AI | perplexity/ |
:white_check_mark: | – | – | – | – |
| Petals | petals/ |
:white_check_mark: | – | – | – | – |
| PG Vector | pg_vector/ |
– | – | – | – | – |
| Pinstripes | pinstripes/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Poe | poe/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Predibase | predibase/ |
:white_check_mark: | – | – | – | – |
| PublicAI | publicai/ |
:white_check_mark: | – | – | – | – |
| Pydantic AI Agents | pydantic_ai_agents/ |
– | – | – | – | – |
| RAGFlow | ragflow/ |
:white_check_mark: | – | – | – | – |
| Recraft | recraft/ |
– | – | :white_check_mark: | – | – |
| Reducto | reducto/ |
– | – | – | – | – |
| Replicate | replicate/ |
:white_check_mark: | – | – | – | – |
| RunwayML | runwayml/ |
– | – | :white_check_mark: | :white_check_mark: | – |
| Sagemaker Chat | sagemaker_chat/ |
:white_check_mark: | – | – | – | – |
| Sambanova | sambanova/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| SAP Generative AI Hub | sap/ |
:white_check_mark: | – | – | – | – |
| Sarvam | sarvam/ |
:white_check_mark: | – | – | – | – |
| Scaleway | scaleway/ |
:white_check_mark: | – | – | – | – |
| SearXNG | searxng/ |
– | – | – | – | – |
| Serper | serper/ |
– | – | – | – | – |
| Snowflake | snowflake/ |
:white_check_mark: | – | – | – | – |
| Soniox | soniox/ |
– | – | – | :white_check_mark: | – |
| Stability AI | stability/ |
– | – | :white_check_mark: | – | – |
| Synthetic | synthetic/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Tavily | tavily/ |
– | – | – | – | – |
| Tencent | tencent/ |
:white_check_mark: | – | – | – | – |
| Tensormesh | tensormesh/ |
:white_check_mark: | – | – | – | – |
| Text Completion Codestral | text-completion-codestral/ |
:white_check_mark: | – | – | – | – |
| Text Completion OpenAI | text-completion-openai/ |
:white_check_mark: | – | – | :white_check_mark: | :white_check_mark: |
| Together AI | together_ai/ |
:white_check_mark: | – | – | – | – |
| Topaz | topaz/ |
– | – | – | – | – |
| Triton | triton/ |
:white_check_mark: | – | – | – | – |
| V0 | v0/ |
:white_check_mark: | – | – | – | – |
| Venice.ai | venice/ |
:white_check_mark: | – | – | – | – |
| Vercel AI Gateway | vercel_ai_gateway/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Vertex AI Agent Engine | vertex_ai/agent_engine/ |
:white_check_mark: | – | – | – | – |
| VLLM | vllm/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Volcengine | volcengine/ |
:white_check_mark: | :white_check_mark: | – | – | – |
| Voyage AI | voyage/ |
– | :white_check_mark: | – | – | – |
| WandB Inference | wandb/ |
:white_check_mark: | – | – | – | – |
| Watsonx Text | watsonx_text/ |
:white_check_mark: | – | – | – | – |
| xAI | xai/ |
:white_check_mark: | – | – | – | – |
| Xiaomi Mimo | xiaomi_mimo/ |
:white_check_mark: | – | – | – | – |
| Xinference | xinference/ |
– | :white_check_mark: | – | – | – |
| Z.AI | zai/ |
:white_check_mark: | – | – | – | – |
165 providers total.
Use any provider by prefixing the model name with the provider’s routing prefix:
from liter_llm import LiterLLM
client = LiterLLM()
# OpenAIresponse = await client.chat("openai/gpt-4o", messages=[ {"role": "user", "content": "Hello!"}])
# Anthropicresponse = await client.chat("anthropic/claude-3-opus", messages=[ {"role": "user", "content": "Hello!"}])
# Groqresponse = await client.chat("groq/llama3-70b", messages=[ {"role": "user", "content": "Hello!"}])Custom Providers
Section titled “Custom Providers”Any OpenAI-compatible API can be used as a custom provider by setting the base URL and API key directly:
response = await client.chat("custom/my-model", base_url="https://my-api.example.com/v1", api_key="my-key", messages=[ {"role": "user", "content": "Hello!"} ])Provider Registry
Section titled “Provider Registry”The full provider registry with base URLs, auth configuration, and model mappings is available at schemas/providers.json.