Qwen: Qwen3.5-122B-A10B

qwen/qwen3.5-122b-a10b

Released Feb 25, 2026262,144 context

$0.40/M input tokens$3.20/M output tokens

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.

Qwen: Qwen3.5-122B-A10B

qwen/qwen3.5-122b-a10b

Released Feb 25, 2026262,144 context

$0.40/M input tokens$3.20/M output tokens

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.

Recent activity on Qwen3.5-122B-A10B

Total usage per day on OpenRouter

Prompt

18.6M

Reasoning

6.43M

Completion

1.1M

Prompt tokens measure input size. Reasoning tokens show internal thinking before a response. Completion tokens reflect total output length.