# Qwen3 Coder Next: AI model fact sheet

- **Provider:** qwen
- **Released:** 2026-02-04
- **Context window:** 262,144 tokens
- **Parameters:** 80B (3B active)
- **API pricing:** $0.20 / 1M input, $1.50 / 1M output
- **OpenRouter ID:** qwen/qwen3-coder-next
- **Capabilities:** conversation, code-generation, agentic-tool-use, tool-use

Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows. It uses a sparse MoE design with 80B total parameters and only 3B activated per token, delivering performance comparable to models with 10 to 20x higher active compute. Operates exclusively in non-thinking mode for streamlined integration.

## Benchmarks

| Benchmark | Score |
| --- | --- |
| SWE-bench Verified | 70.6% |

Source: real side-by-side outputs, pricing and specs at https://rival.tips/models/qwen3-coder-next