ScoutCode
← All models

Speed Router

RouterFreeToolsVision
speed-router

Rates how hard each request is, then picks Qwen 3.8 27B, Muse Spark 1.3 Contributor or GPT 6 Luna and the reasoning effort to match.

Cost
What the answering model costs
Context
1M tokens
Price
Free
Status
Available

Routes to

Tried in this order. If one fails or is rate limited, the next one answers.

  1. Qwen 3.8 27B qwen3.8-27b
  2. Muse Spark 1.3 Contributor muse-spark-1.3-contributor
  3. GPT 6 Luna gpt-6-luna

Use it

Send "model": "speed-router" to the OpenAI-compatible endpoint. API docs

curl http://localhost:4321/v1/chat/completions \
  -H "Authorization: Bearer $SCOUTCODE_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "speed-router",
    "messages": [{ "role": "user", "content": "Hello!" }]
  }'