---
id: "qwen/qwen3.5-122b-a10b"
also: ["qwen/qwen3.5-122b-a10b-20260224", "Qwen/Qwen3.5-122B-A10B", "hf/Qwen/Qwen3.5-122B-A10B"]
cheapest_list_usd_per_m: "0.26 in, 2.08 out, at alibaba on openrouter"
cheapest_effective_usd_per_m: "0.2597 in, 2.0799 out, at siliconflow/fp8 on openrouter"
context: 262144
max_output: 65536
input: ["text", "image", "video"]
output: ["text"]
released: "2026-02-25"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logit_bias", "logprobs", "max_tokens", "min_p", "presence_penalty", "reasoning", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 7
gateways: "openrouter (5), hf (2)"
no_training: 5
trains: 0
retains_prompts: 2
coding: 45.7000
agentic: 7.6000
---
# Qwen: Qwen3.5-122B-A10B

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...