Where do model IDs and prices come from?

Render them from GET https://llm.chutes.ai/v1/models; examples can go stale.

Can default:latency be used without setup?

No. Saved default aliases require a dashboard Model Routing pool. Inline pools and concrete live model IDs are zero setup.

Source draft Connect quickstarts

Catalog checked Jun 25, 2026

Claude + Chutes recipes

Plugin install, credential checks, MCP tools, project rules, model picking, and editor portability for Claude Code or Cowork.

Install the Claude plugin

Use the marketplace flow for the full nine-skill suite.

Source

Install the Claude plugin

text

/plugin marketplace add Veightor/chutes-agent-toolkit
/plugin install chutes-ai@chutes-agent-toolkit

Check credential storage

Verify the keychain profile without printing secrets, then export CHUTES_API_KEY for local scripts.

Source

Check credential storage

bash

python plugins/chutes-ai/skills/chutes-ai/scripts/manage_credentials.py check
export CHUTES_API_KEY="$(python plugins/chutes-ai/skills/chutes-ai/scripts/manage_credentials.py get --field api_key)"

Add project-level Chutes rules

Use this in CLAUDE.md when a repo should remember Chutes endpoint, auth, routing, and live-catalog rules.

Source

Add project-level Chutes rules

markdown

## Chutes inference
- Base URL: https://llm.chutes.ai/v1
- Auth: Authorization: Bearer $CHUTES_API_KEY
- Never send X-API-Key; inference ignores it.
- Never hardcode model IDs, prices, context, features, or TEE claims.
- Discover models with GET https://llm.chutes.ai/v1/models.
- default:* aliases need a dashboard Model Routing pool; concrete IDs and inline pools work with zero setup.

Install MCP tools for Claude Desktop

Read tools are verified; write/deploy tools keep [BETA] labels.

Source

Install MCP tools for Claude Desktop

bash

uv tool install chutes-mcp-server \
  --from plugins/chutes-ai/skills/chutes-mcp-portability/mcp-server

Which lane?

The routing pool below is computed from the live catalog for agentic work. Use it inline for zero setup, or configure a saved pool before using default:* aliases.

Concrete model

Qwen/Qwen3-32B-TEE

Use when

You need a specific context window, feature set, modality, or price.

Inline pool

Qwen/Qwen3-32B-TEE,google/gemma-4-31B-turbo-TEE,MiniMaxAI/MiniMax-M2.5-TEE:latency

Use when

You want failover or latency selection without saved alias setup.