Qwen2.5-Omni-7B
Alibaba Cloud / Qwen TeamQwenOpen WeightApache 2.0 · Commercial OK
Description
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.
Release Date
2025-03-27
Parameters
7.0B
Context Length
33K
Modalities
audio, image, text, video
Capability Radar
60
general
80
coding
60
reasoning
26
scienceest.
66
agents
80
multimodal
Science is estimated from LLM Stats science scores or reasoning when no dedicated science benchmarks are available.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Multimodal Ranking | 144 | 22.0 | LS |
Benchmark Scores (LLM Stats)
(LLM Stats (zeroeval))Audio
VocalSound
93.9%SR
GiantSteps Tempo
88.0%SR
MMAU Music
69.2%SR
MMAU Sound
67.9%SR
MMAU
65.6%SR
MMAU Speech
59.8%SR
CoVoST2 en-zh
0.41 / 100SR
MusicCaps
32.8%SR
Common Voice 15
0.08 / 100SR
Chat
VoiceBench Avg
74.1%SR
MM-MT-Bench
0.06 / 100SR
General
MultiPL-E
65.8%SR
NMOS
0.05 / 100SR
Healthcare
CRPErelation
76.5%SR
Language
FLEURS
95.9%SR
MMLU-Redux
71.0%SR
MMLU-Pro
47.0%SR
Math
GSM8k
88.7%SR
MATH
71.5%SR
MathVista
67.9%SR
LiveBench
29.6%SR
MathVision
25.0%SR
Multimodal
MMMU
59.2%SR
Meld
57.0%SR
OmniBench
56.1%SR
OmniBench Music
52.8%SR
Reasoning
ChartQAMasry et al. (2022)
85.3%SR
HumanEvalOpenAI (2021)
78.7%SR
MBPP
0.73 / 100SR
GPQANYU + Cohere + Anthropic (2023)
30.8%SR
Vision
DocVQADocVQA (2020)
95.2%SR
TextVQA
84.4%SR
AI2D
83.2%SR
MMBench-V1.1
81.8%SR
VideoMME w sub.
72.4%SR
RealWorldQA
70.3%SR
MVBench
70.3%SR
EgoSchema
68.6%SR
PointGrounding
66.5%SR
MMStar
64.0%SR
MME-RealWorld
61.6%SR
MuirBench
59.2%SR
OCRBench_V2
57.8%SR
ODinW
42.4%SR
MMMU-Pro
36.6%SR
AA Evaluation Indices
(Artificial Analysis)No AA evaluation data available
LLM Stats Category Scores
(LLM Stats (zeroeval))Speech To Text90
Image To Text80
Code80
Language70
Long Context70
Reasoning70
Safety70
Spatial Reasoning70
Audio70
Video70
Vision70
Math60
Multimodal60
General60
Legal50
Finance50
Healthcare50
Chat40
Communication40
Physics30
Biology30
Chemistry30
Pricing
Input Price$0.1 / 1M tokens
Output Price$0.4 / 1M tokens
Blended Price (3:1)$0.175 / 1M tokens
Speed
No speed data available
Provider Price Ranking
Provider Price Ranking
3 providers
Cheapest: Alibaba (China)Most Expensive: Alibaba
ProviderInputOutput
1Alibaba (China)Cheapest
$0.087
$0.345
2Alibaba Cloud / Qwen TeamPRIMARY
$0.1
$0.4
3Alibaba
$0.1
$0.4
Compare pricing across different API providers for this model.