# Qwen3 235B A22B

Qwen3-235B-A22B is a 235B parameter mixture-of-experts (MoE) model developed by Qwen, activating 22B parameters per forward pass. It supports seamless switching between a "thinking" mode for complex reasoning, math, and code tasks, and a "non-thinking" mode for general conversational efficiency. The model demonstrates strong reasoning ability, multilingual support (100+ languages and dialects), advanced instruction-following, and agent tool-calling capabilities. It natively handles a 32K token context window and extends up to 131K tokens using YaRN-based scaling.

## Model Information

- **Organization**: [qwen](/llm.txt)
- **Slug**: qwen3-235b-a22b
- **Available at Providers**: 26
- **Release Date**: April 29, 2025

### Benchmark Scores
- Weekly: 0.73
- AIME 2025: 0.815
- GPQA: 0.4747

## Providers

| Provider | Name | $ Input (per 1M) | $ Output (per 1M) | Free | Link |
|----------|------|-----------------|------------------|------|------|
| [AIHubMix](/llm/aihubmix.txt) | qwen3-235b-a22b | 0.28 | 1.12 |  | [View](https://aihubmix.com/model/qwen3-235b-a22b) |
| [FastRouter](/llm/fastrouter.txt) | Qwen: Qwen3 235B A22B | 0.13 | 0.60 |  | [View](https://fastrouter.ai/models/qwen/qwen3-235b-a22b) |
| [Alibaba](/llm/alibaba.txt) | Qwen3 235B-A22B | 0.70 | 2.80 |  |  |
| [Nvidia](/llm/nvidia.txt) | qwen3-235b-a22b | 0.00 | 0.00 | Yes | [View](https://build.nvidia.com/qwen/qwen3-235b-a22b) |
| [Alibaba (China)](/llm/alibabacn.txt) | Qwen3 235B-A22B | 0.29 | 1.15 |  |  |
| [SiliconFlow](/llm/siliconflow.txt) | Qwen/Qwen3-235B-A22B | 0.35 | 1.42 |  |  |
| [Nano-GPT](/llm/nanogpt.txt) | Qwen 3 235b A22B |  |  |  |  |
| [OpenRouter](/llm/openrouter.txt) | Qwen3 235B A22B | 0.46 | 1.82 |  | [View](https://openrouter.ai/qwen/qwen3-235b-a22b-04-28) |
| [Requesty](/llm/requesty.txt) |  | 0.20 | 0.60 |  |  |
| [ValorGPT](/llm/valorgpt.txt) | Qwen3 235B A22B |  |  |  | [View](https://www.valorgpt.com/models/qwen-qwen3-235b-a22b) |
| [ValorGPT](/llm/valorgpt.txt) | Qwen3 235B A22B |  |  | Yes | [View](https://www.valorgpt.com/models/qwen-qwen3-235b-a22b-free) |
| [Yupp](/llm/yupp.txt) | Qwen 3 235B A22B (OpenRouter) |  |  |  |  |
| [Yupp](/llm/yupp.txt) | Qwen3-235B-A22B-Instruct-2507 (OpenRouter) |  |  |  |  |
| [LangDB](/llm/langdb.txt) | qwen3-235b-a22b |  |  |  | [View](https://langdb.ai/app/models) |
| [302.AI](/llm/302ai.txt) | Qwen3-235B-A22B | 0.29 | 2.86 |  |  |
| [Kilo Code](/llm/kilocode.txt) | Qwen: Qwen3 235B A22B | 0.30 | 1.20 |  |  |
| [Baidu AI Studio](/llm/baidu.txt) |  |  |  |  | [View](https://aistudio.baidu.com) |
| [Blackbox AI](/llm/blackboxai.txt) | Qwen: Qwen3 235B A22B | 0.13 | 0.60 |  |  |
| [Helicone](/llm/helicone.txt) | Qwen3 235B A22B Thinking | 0.30 | 2.90 |  | [View](https://www.helicone.ai/model/qwen3-235b-a22b-thinking) |
| [Vercel AI Gateway](/llm/vercel.txt) | Qwen3 235B A22B Thinking 2507 | 0.30 | 2.90 |  |  |
| [Arena AI](/llm/arenaai.txt) |  |  |  |  |  |
| [NetMind](/llm/netmind.txt) | qwen-3-235b | 0.20 | 0.80 |  |  |
| [CometAPI](/llm/cometapi.txt) |  | 0.34 | 1.34 |  |  |
| [Qiniu](/llm/qiniuai.txt) | Qwen 3 235B A22B |  |  |  |  |
| [ApiYI](/llm/apiyi.txt) | qwen3-235b-a22b |  |  |  |  |
| [WaveSpeed AI](/llm/wavespeed.txt) | qwen3-235b-a22b | 0.22 | 0.66 |  |  |

---

[← Back to all providers](/llm.txt)