Better than 81% of all models
Context
N/A
Input $/1M
TBD
Output $/1M
TBD
Type
image-text-to-text
License
Open Source
Benchmarks
11 tested
Data updated today
About
Qwen image text to text model. 1523K downloads on HuggingFace.
Tested on 11 benchmarks with 47.3% average. Top scores: JSQuAD (89.9%), JCommonsenseQA (87.8%), JNLI (74.4%).
Capabilities
reasoning
13.6
#115 globally
math
19.9
#162 globally
knowledge
21.8
#192 globally
language
67.9
#73 globally
general
35.9
#23 globally
Benchmark Scores
Compare AllTested on 11 benchmarks · Ranked across 5 categories
Score Distribution (all 231 models)
0255075100
▲ You are here
reasoningCompare reasoning →
MUSR
13.6—HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.
mathCompare math →
MATH Level 5
19.9—HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.
knowledgeCompare knowledge →
MMLU-PRO
34.4—HuggingFace MMLU-Pro. Harder version of MMLU with 10 answer choices instead of 4 and more challenging questions.
GPQA
9.3—HuggingFace evaluation of GPQA (Graduate-Level Google-Proof Q&A). PhD-level science questions that cannot be easily searched.
vs GPT-5.1-Codex-Mini
vs Grok 3 Beta
vs Gemini 2.5 Pro Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Links
Info
Research
Documentation
Community
Source Code
BenchGecko API
qwen-qwen2-vl-7b-instruct
Specifications
- Typeimage-text-to-text
- ContextN/A
- ReleasedAug 2024
- LicenseOpen Source
- StatusActive
Available On
AlibabaTBDLearn More
Share & Export
Frequently Asked Questions
Qwen2 VL 7B Instruct is an open-source image-text-to-text AI model by Alibaba, released in August 2024. It has an average benchmark score of 67.6.