Powered byLightning AI
← Managed inference

Qwen3 Coder

Coding agents and IDE workflows at repo-scale context

Start free

Overview

Qwen3-Coder-Next is built specifically for coding agents and local development: 80B total parameters with 3B active, which Qwen says reaches performance comparable to models with 10-20x more active parameters. Its training recipe targets long-horizon reasoning, complex tool use and recovery from execution failures, and its long context plus scaffold adaptability let it slot into CLI and IDE harnesses such as Claude Code, Qwen Code, Cline and Trae.

Strengths

  • Trained for long-horizon agentic coding
  • Recovers from failed tool executions
  • Drops into existing CLI and IDE harnesses

Use cases

  • Multi-file refactors and repo-scale edits
  • Agentic coding loops driven by test feedback
  • Code review and framework migrations

Good to know

Non-thinking mode only, and specialised for code -- reach for a general chat model for open-ended writing or analysis. Served with a 131K context window, below the 256K the model card advertises.

API access

One OpenAI-compatible API for every model in the catalog, plus the CLI and the function SDK inside workflows. Sign up free, add a card for $5 in credits, and these requests work:

qwen3-coder · curl
curl https://gateway.graphn.ai/v1/chat/completions \
  -H "Authorization: Bearer $GRAPHN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3-coder",
    "messages": [{"role": "user", "content": "Hello"}]
  }'