# Qwen3 235B A22B: AI model fact sheet

- **Provider:** qwen
- **Released:** 2025-04-28
- **Context window:** 32,768 tokens
- **Parameters:** 235B (22B Active)
- **OpenRouter ID:** qwen/qwen3-235b-a22b:free
- **Capabilities:** conversation, reasoning, code-generation, analysis

Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model from Alibaba's Qwen team, activating 22B parameters per forward pass. Features seamless switching between 'thinking' mode (complex tasks) and 'non-thinking' mode (general conversation). Strong reasoning, multilingual (100+), instruction-following, and tool-calling. 32K context, extendable to 131K.

## Benchmarks

| Benchmark | Score |
| --- | --- |
| Codeforces | Competitive with o3-mini/Gemini 2.5 Pro |
| AIME | Competitive with o3-mini |
| SWE-bench Verified | 61.7% |
| LiveCodeBench | 70.7% |

Source: real side-by-side outputs, pricing and specs at https://rival.tips/models/qwen3-235b-a22b