跳转到主要内容

Qwen3.5 0.8B (Non-reasoning)

AlibabaQwen开源权重Apache 2.0 · 商用许可

描述

Qwen3.5-0.8B is a 0.8 billion parameter vision-language model using Gated DeltaNet hybrid architecture with a 3:1 ratio of linear attention to full softmax attention. It supports 262K native context length and features both thinking and non-thinking modes.

发布日期
2026-03-02
参数规模
800M
上下文长度
131K
支持模态
text

能力雷达图

30
general
0
coding
40
reasoning
9
science估算
20
agents
0
multimodal

缺少专门科学评测时,Science 由 LLM Stats 科学得分或推理能力估算。

排行榜排名

领域#排名分数来源
智能体能力模型榜146
27.0
LS

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Agents

t2-bench11.6%自报

Chat

IFEvalGoogle Research (2023)44.0%自报
Multi-Challenge18.9%自报

General

C-Eval50.5%自报
NOVA-6342.4%自报
Include40.6%自报
MAXIFE39.2%自报

Instruction Following

IFBench21.0%自报

Language

MMLU-Redux59.5%自报
MMMLU44.3%自报
MMLU-Pro42.3%自报
MMLU-ProX34.6%自报
WMT24++27.2%自报

Long Context

LongBench v226.1%自报

Math

PolyMATH8.2%自报

Reasoning

Global PIQA59.4%自报
SuperGPQA21.3%自报
GPQANYU + Cohere + Anthropic (2023)11.9%自报
AA-LCR4.7%自报

Tool Calling

BFCL-V425.3%自报

AA 评测指数

(Artificial Analysis)

暂无 AA 评测数据

LLM Stats 分类评分

(LLM Stats (zeroeval))
Language
40
Math
40
Structured Output
40
Chat
30
Instruction Following
30
Legal
30
Physics
30
Reasoning
30
Finance
30
General
30
Healthcare
30
Long Context
20
Agents
20
Chemistry
20
Communication
20
Economics
20
Tool Calling
20
Multimodal
10
Spatial Reasoning
10
Biology
10
Vision
10

定价

输入价格$0.18 / 1M tokens
输出价格$0.7 / 1M tokens
混合价格(3:1)$0.31 / 1M tokens

速度

暂无速度数据

供应商价格排行

供应商价格排行

1 个供应商

供应商输入输出
1Alibaba主要
$0.18
$0.7

比较该模型在不同 API 供应商之间的定价。

外部链接