BlockRun
Get started
Integrations / Continue

Point Continue's OpenAI provider at BlockRun and code with any model

Set Continue's OpenAI provider apiBase to the blockrun-litellm sidecar and chat, edit and autocomplete can use any catalog model — one YAML block per model, no provider key.

Get your agent on BlockRun 

How it works

Continue configures models in a YAML file, and a model block with provider openai and an apiBase is enough to reach any OpenAI-compatible endpoint. The blockrun-litellm sidecar is one that signs an x402 payment per request from your wallet, so Continue's chat, edit and agent modes can name any catalog model with no provider key.

Continue also loads MCP servers, and the BlockRun MCP entry adds the media and data tools — image generation, web search, market data — to the agent mode. Use the provider block for the model and the MCP block for the tools; both pay from the same wallet.

Behind it is the BlockRun catalog: 78 chat models plus image, video, music and speech generation, search, market data and multi-chain RPC, priced per call. 6 open-weight models are free and need no wallet.

Setup

1. Start the sidecar

pip install 'blockrun-litellm[proxy]'\nexport BLOCKRUN_WALLET_KEY=0x<base wallet key>\nblockrun-litellm-proxy --port 4001

2. ~/.continue/config.yaml

models:
  - name: Claude Sonnet 5 via BlockRun
    provider: openai
    model: anthropic/claude-sonnet-5
    apiBase: http://127.0.0.1:4001/v1
    apiKey: dummy
    roles: [chat, edit]

Questions

How do I add BlockRun to Continue?
Start the sidecar and add a model block with provider openai, the sidecar as apiBase and a catalog id as the model, as in the setup. It appears in Continue's model picker.
Can I use BlockRun for autocomplete as well as chat?
Yes. Give the model block the autocomplete role, or add a second block with a fast, cheap catalog model for it.
Does Continue's agent mode get BlockRun's tools?
Add the BlockRun MCP server to Continue's MCP configuration and the agent mode can generate images, search the web and read market data through it.
Which model ids do I use in config.yaml?
The catalog ids as they appear on the models page — the sidecar takes them bare, without a prefix.
Continue on BlockRun vs Continue with a provider key?
Same assistant, same config format. One wallet replaces a key per provider, and every model in the catalog is one YAML block away.
What does a call cost through this integration?
The provider's published token rate with no markup on tokens, plus a flat per-call fee, quoted in dollars in the 402 before the call runs. Image, video and speech calls are priced per unit the same way. The live rate card is on the models page.
Do I need an API key or an account?
No. On the pay-per-call door the request carries its own payment from a wallet the SDK or MCP server creates on first run; there is nothing to sign up for. Teams that prefer keys get an API key and a monthly invoice instead.

Related