# Qwen: Qwen3.5 122B A10B: AI model fact sheet

- **Provider:** qwen
- **Released:** 2026-02-25
- **Context window:** 262,144 tokens
- **Parameters:** 122B (10B active)
- **API pricing:** $0.40 / 1M input, $3.20 / 1M output
- **OpenRouter ID:** qwen/qwen3.5-122b-a10b
- **Capabilities:** conversation, reasoning, code-generation, analysis, tool-use, agentic-tool-use, translation

The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of overall performance, this model is second only to Qwen3.5-397B-A17B. Its text capabilities significantly outperform those of Qwen3-235B-2507, and its visual capabilities surpass those of Qwen3-VL-235B.

## Benchmarks

| Benchmark | Score |
| --- | --- |
| MMLU-Pro | 86.7 |
| GPQA Diamond | 86.6% |
| SWE-bench Verified | 72.0% |
| LiveCodeBench v6 | 78.9 |
| HMMT Feb 25 | 91.4% |
| IFEval | 93.4% |
| MMMU | 83.9 |
| BFCL-V4 | 72.2 |

Source: real side-by-side outputs, pricing and specs at https://rival.tips/models/qwen3.5-122b-a10b