Model Providers Overview

Continue supports a wide range of AI model providers to power different features like chat, code editing, autocompletion, and embeddings. This overview helps you navigate through the available options and find the right provider for your needs.

Popular Model Providers

These are the most commonly used model providers that offer a wide range of capabilities:

Provider	Description	Capabilities
Anthropic	Providers of Claude models, known for long context windows and strong reasoning	Chat, Edit, Apply, Embeddings
OpenAI	Creators of GPT models with strong coding capabilities	Chat, Edit, Apply, Embeddings
Azure	Microsoft’s cloud platform offering OpenAI models	Chat, Edit, Apply, Embeddings
Amazon Bedrock	AWS service offering access to various foundation models	Chat, Edit, Apply, Embeddings
Ollama	Run open-source models locally with a simple interface	Chat, Edit, Apply, Embeddings, Autocomplete
Google Gemini	Google’s multimodal AI models	Chat, Edit, Apply, Embeddings
DeepSeek	Specialized code models with strong performance	Chat, Edit, Apply
Mistral	High-performance open models with commercial offerings	Chat, Edit, Apply, Embeddings
xAI	Grok models from xAI	Chat, Edit, Apply
Vertex AI	Google Cloud’s machine learning platform	Chat, Edit, Apply, Embeddings
Inception	On-premises open-source model runners	Chat, Edit, Apply
HuggingFace	Platform for open source models with inference providers and endpoints	Chat, Edit, Apply, Embeddings

Additional Model Providers

Beyond the top-level providers, Continue supports many other options:

Hosted Services

Provider	Description
Groq	Ultra-fast inference for various open models
Together AI	Platform for running a variety of open models
DeepInfra	Hosting for various open source models
OpenRouter	Gateway to multiple model providers
ClawRouter	Open-source LLM router with automatic cost-optimized model selection
Tetrate Agent Router Service	Gateway with intelligent routing across multiple model providers
Cohere	Models specialized for semantic search and text generation
NVIDIA	GPU-accelerated model hosting
Cloudflare	Edge-based AI inference services
MiniMax	High-performance models with 200K+ context window

Local Model Options

Provider	Description
LM Studio	Desktop app for running models locally
llama.cpp	Optimized C++ implementation for running LLMs
LlamaStack	Stack for running Llama models locally
llamafile	Self-contained executable model files

Enterprise Solutions

Provider	Description
SambaNova	Enterprise AI platform
Watson x	IBM’s enterprise AI platform
Sagemaker	AWS machine learning platform
Nebius	Cloud-based machine learning platform

How to Choose a Model Provider

When selecting a model provider, consider:

1. Hosting preference: Do you need local models for offline use or privacy, or are you comfortable with cloud services?
2. Performance requirements: Different providers offer varying levels of speed, quality, and context length.
3. Specific capabilities: Some models excel at code generation, others at embeddings or reasoning tasks.
4. Pricing: Costs vary significantly between providers, from free local options to premium cloud services.
5. API key requirements: Most cloud providers require API keys that you’ll need to configure.

Configuration Format

You can add models to your config.yaml file like this:

models:
  - name: Claude 4 Sonnet
    provider: anthropic # Choose provider from the lists above
    model: claude-sonnet-4-20250514 # Specific model name
    apiKey: ${{ secrets.OPENAI_API_KEY }}
    roles:
      - chat
      - edit
      - apply

For more detailed configuration, visit the specific provider pages linked above.

Change Your Model Provider

Continue allows you to choose your favorite or even add multiple model providers. This allows you to use different models for different tasks, or to try another model if you’re not happy with the results from your current model. Continue supports all of the popular model providers, including OpenAI, Anthropic, Microsoft/Azure, Mistral, and more. You can even self host your own model provider if you’d like. Learn more about model providers.