Model details

Qwen3.5 122B-A10B

Qwen · qwen3.5-122b-a10b

The Qwen3.5 series Qwen3.5 122B-A10B native vision-language model is designed based on a hybrid architecture, combining a linear attention mechanism with a sparse mixture-of-experts model to achieve higher inference efficiency. The overall performance of this model is second only to Qwen3.5-397B-A17B, with text capabilities significantly superior to Qwen3-235B-2507 and visual capabilities superior to Qwen3-VL-235B.

能读 262K能看图能调工具会深度思考
Official price¥0.8000/ ¥6.400每百万 Token
中转最低76% of official¥0.6120 / ¥4.896
108家在卖1份验真报告

The best way to buy Qwen3.5 122B-A10B

The best deal for each of three ways to buy. Tap a card to see every option of that kind.

Pay-as-you-go API · relay stationsFor coding and tools like Cursor and Claude Code
For using the web app and mobile app, no coding
Official subscriptionsFor a genuine official plan in your own name

Qwen3.5 122B-A10B

The Qwen3.5 series Qwen3.5 122B-A10B native vision-language model is designed based on a hybrid architecture, combining a linear attention mechanism with a sparse mixture-of-experts model to achieve higher inference efficiency. The overall performance of this model is second only to Qwen3.5-397B-A17B, with text capabilities significantly superior to Qwen3-235B-2507 and visual capabilities superior to Qwen3-VL-235B.

View intro source
Official price · Input¥0.8000
Official price · Output¥6.400
Sellers108 家
Public verification reports1 份

The best way to buy Qwen3.5 122B-A10B

The best deal for each of three ways to buy. Tap a card to see every option of that kind.

Pay-as-you-go API · relay stationsFor coding and tools like Cursor and Claude Code
For using the web app and mobile app, no coding
Official subscriptionsFor a genuine official plan in your own name
No reviews yet — write the first oneSee reviews ↓

模型介绍

The Qwen3.5 series Qwen3.5 122B-A10B native vision-language model is designed based on a hybrid architecture, combining a linear attention mechanism with a sparse mixture-of-experts model to achieve higher inference efficiency. The overall performance of this model is second only to Qwen3.5-397B-A17B, with text capabilities significantly superior to Qwen3-235B-2507 and visual capabilities superior to Qwen3-VL-235B.

模型基础信息

厂商
通义千问
模型类型
文本与多模态
一次能读多少
262K Token一次对话能塞进去的最大内容量,长文档、长代码要看这项
一次能写多少
65,536 Token单次回答的长度上限
能接收什么
图像 · 文本 · 视频可以发给它的内容类型
能产出什么
文本它能生成的内容类型
能看图吗
可以
能调工具吗
可以能不能自动调用外部函数、查资料、执行动作
怎么计费
按 Token 计费官方按什么单位收钱
现在还能用吗
可以用 · 状态来源阿里云 Model Studio 官方模型目录核对
这些资料是哪个地区的资料适用地区
中国大陆厂商专门为中国大陆公布的资料。
它会做什么能读上传的文件会聊天擅长推理支持结构化输出可以设定角色回答风格可调会写字会调工具会看图

官方价格

输入 · 中国大陆
0.8 CNY每 100 万 Token
输出 · 中国大陆
6.4 CNY每 100 万 Token

中国大陆 / 全球 · 按适用条件分档计价

折扣与价格评分按中国大陆的官方价计算

查看分档价格

输入

不分档2.672 CNY每百万 Token
输入 128,000 Token 以内0.8 CNY每百万 Token
输入 256,000 Token 以内,输入 128,000 Token 以上2 CNY每百万 Token

输出

不分档21.37 CNY每百万 Token
输入 128,000 Token 以内6.4 CNY每百万 Token
输入 256,000 Token 以内,输入 128,000 Token 以上16 CNY每百万 Token

中国大陆 / 全球 · 采集于 2026-11-18 · 查看价格来源

它还叫这些名字(15 条)
Qwen3.5-122B-A10Bqwen3.5-122b-a10b-highqwen3.5-122b-a10b-lowqwen3.5-122b-a10b-mediumqwen3.5-122b-a10b-minimalqwen3.5-122b-a10b-non-thinkingqwen3.5-122b-a10b-nothinkingqwen3.5-122b-a10b-thinkingqwen3.5-122b-a10b-thinking-highqwen3.5-122b-a10b-thinking-lowqwen3.5-122b-a10b-thinking-mediumqwen3.5-122b-a10b-thinking-minimalqwen3.5-122b-a10b-thinking-xhighqwen3.5-122b-a10b-xhigh通义千问-3.5-122b-a10b

不同地方写法可能不一样,列在这里方便你对上号。名字太模糊、认不准的不会公开。

User reviews

No reviews yet. If you have used it, share the first honest review and help others avoid pitfalls.

Review this model

Rating
Optional
编辑器加载中…