跳转到主要内容

Qwen2.5-Omni-7B

Alibaba Cloud / Qwen TeamQwenOpen WeightApache 2.0 · Commercial OK

描述

Qwen2.5-Omni is the flagship end-to-end multimodal model in the Qwen series. It processes diverse inputs including text, images, audio, and video, delivering real-time streaming responses through text generation and natural speech synthesis using a novel Thinker-Talker architecture.

发布日期
2025-03-27
参数规模
7.0B
上下文长度
支持模态

能力雷达图

50
general
80
coding
60
reasoning
26
science估算
0
agents
90
multimodal

Science 在缺少专门科学评测时使用推理能力代理估算。

排行榜排名

领域#排名分数来源
多模态榜52
74.0
LS

基准测试分数 (LLM Stats)

Audio

VocalSound93.9%自报
GiantSteps Tempo88.0%自报
MMAU Music69.2%自报
MMAU Sound67.9%自报
MMAU65.6%自报
MMAU Speech59.8%自报
OmniBench Music52.8%自报
CoVoST2 en-zh0.41 / 100自报
MusicCaps32.8%自报
Common Voice 150.08 / 100自报

Biology

GPQA30.8%自报

Code

HumanEval78.7%自报

Communication

VoiceBench Avg74.1%自报
MM-MT-Bench0.06 / 100自报

Creativity

Meld57.0%自报

Finance

MMLU-Pro47.0%自报

General

MBPP0.73 / 100自报
MMLU-Redux71.0%自报
MultiPL-E65.8%自报
MMStar64.0%自报
MME-RealWorld61.6%自报
MMMU59.2%自报
MMMU-Pro36.6%自报
LiveBench29.6%自报
NMOS0.05 / 100自报

Grounding

PointGrounding66.5%自报

Healthcare

CRPErelation76.5%自报

Image To Text

DocVQA95.2%自报
TextVQA84.4%自报
OCRBench_V257.8%自报

Language

FLEURS0.04 / 100自报

Long Context

EgoSchema68.6%自报

Math

GSM8k88.7%自报
MATH71.5%自报
MathVista67.9%自报
MathVision25.0%自报

Multimodal

ChartQA85.3%自报
AI2D83.2%自报
MMBench-V1.181.8%自报
VideoMME w sub.72.4%自报
MVBench70.3%自报
MuirBench59.2%自报
OmniBench56.1%自报

Spatial Reasoning

RealWorldQA70.3%自报

Vision

ODinW42.4%自报

AA 评测指数

暂无 AA 评测数据

LLM Stats 分类评分

Image To Text
90
Code
80
Spatial Reasoning
70
Video
70
Vision
70
Long Context
70
Math
60
Multimodal
60
Reasoning
60
Finance
50
General
50
Healthcare
50
Language
50
Legal
50
Biology
30
Chemistry
30
Physics
30
Communication
10
Speech To Text
0

定价

暂无定价数据

速度

暂无速度数据

可用提供商

(LS 内部计价单位)

暂无提供商数据

外部链接