Alibaba Cloud / Qwen Team model product
Qwen2.5 VL 7B
5-VL is a vision-language model from the Qwen family.
Updated Aug 12, 2026. Default version: Qwen2.5 VL 7B Instruct
LLMBoard score12.8Qwen2.5 VL 7B Instruct
Coverage20%28 benchmark families
Context windowN/ATokens
Official input price$0.35Alibaba API
Qwen2.5 VL 7B Specifications
Technical details for the model's default version.
- Version
- Qwen2.5 VL 7B Instruct
- Released
- Jan 26, 2025
- Knowledge cutoff
- Unknown
- Parameters
- 8.3B
- Context window
- N/A
- Max output
- N/A
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Apache 2.0
Qwen2.5 VL 7B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Qwen2.5 VL 7B Benchmark Results
Benchmark scores for Qwen2.5 VL 7B Instruct.
| MobileMiniWob++_SR | 91.4% | 01 | 2 | 100.0% | C | |
| DocVQA | 95.7% | 02 | 26 | 96.0% | C | |
| InfoVQA | 82.6% | 02 | 9 | 87.5% | C | |
| MMT-Bench | 63.6% | 02 | 4 | 66.7% | C | |
| MMVet | 67.1% | 02 | 2 | 0.0% | C | |
| PerceptionTest | 70.5% | 02 | 2 | 0.0% | C | |
| TempCompass | 71.7% | 02 | 2 | 0.0% | C | |
| TextVQA | 84.9% | 02 | 15 | 92.9% | C | |
| AITZ_EM | 81.9% | 03 | 3 | 0.0% | C | |
| Android Control High_EM | 60.1% | 03 | 3 | 0.0% | C | |
| Android Control Low_EM | 91.4% | 03 | 3 | 0.0% | C | |
| MMBench-Video | 1.8% | 03 | 3 | 0.0% | C | |
| LongVideoBench | 54.7% | 04 | 4 | 0.0% | C | |
| MMBench | 84.3% | 05 | 9 | 50.0% | C | |
| AndroidWorld_SR | 25.5% | 07 | 8 | 14.3% | C | |
| ChartQA | 87.3% | 09 | 24 | 65.2% | C | |
| MLVU | 70.2% | 10 | 10 | 0.0% | C | |
| VideoMME w/o sub. | 65.1% | 10 | 10 | 0.0% | C | |
| VideoMME w sub. | 71.6% | 10 | 10 | 0.0% | C | |
| CharadesSTA | 43.6% | 12 | 12 | 0.0% | C | |
| CC-OCR | 77.8% | 13 | 18 | 29.4% | C | |
| MVBench | 69.6% | 13 | 17 | 25.0% | C | |
| OCRBench | 86.4% | 14 | 22 | 38.1% | C | |
| ScreenSpot | 84.7% | 15 | 16 | 6.7% | C | |
| Hallusion Bench | 52.9% | 16 | 16 | 0.0% | C | |
| MMStar | 63.9% | 19 | 22 | 14.3% | C | |
| MathVista-Mini | 68.2% | 20 | 23 | 13.6% | C | |
| LVBench | 45.3% | 22 | 24 | 8.7% | C | |
| ScreenSpot Pro | 29.0% | 25 | 25 | 0.0% | C | |
| MathVision | 25.1% | 31 | 32 | 3.2% | C | |
| MMMU | 58.6% | 49 | 63 | 22.6% | C | |
| MMMU-Pro | 38.3% | 64 | 66 | 3.1% | C |
Qwen2.5 VL 7B Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
Qwen2.5 VL 7B Runtime Performance
Provider-specific output speed and catalog latency for Qwen2.5 VL 7B Instruct. Runtime does not affect the capability score.
No runtime data
Browse runtime rankingsNo provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Qwen2.5 VL 7B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.35 input, $1.05 output per 1M
- Official provider
- Alibaba
- Lowest third-party
- N/A
- Tracked offerings
- 3
| Alibaba (China) | qwen2-5-vl-7b-instruct | global | $0.287 | $0.717 | 131.1K | |
| Alibaba | qwen2-5-vl-7b-instruct | global | $0.35 | $1.05 | 131.1K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Qwen2.5 VL 7B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Qwen2.5 VL 7B Instruct | 12.8 | 8.3B | N/A | N/A | No | Apache 2.0 |
What is Qwen2.5 VL 7B?
Key information about Qwen2.5 VL 7B and its available data.
5-VL is a vision-language model from the Qwen family.
Key enhancements include visual understanding (objects, text, charts, layouts), visual agent capabilities (tool use, computer/phone control), long video comprehension with event pinpointing, visual localization (bounding boxes/points), and structured output generation.
Data as of 2026-08-11.
Qwen2.5 VL 7B vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Qwen2.5 VL 7B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AC
Qwen2.5 32B
Alibaba Cloud / Qwen Team
AC
QvQ 72B
Alibaba Cloud / Qwen Team
AC
Qwen2.5 Omni 7B
Alibaba Cloud / Qwen Team
AC
Qwen2 72B
Alibaba Cloud / Qwen Team
AC
Qwen2.5 14B
Alibaba Cloud / Qwen Team
AC
Qwen2.5 VL 32B
Alibaba Cloud / Qwen Team
FAQ
Common questions about Qwen2.5 VL 7B.
When was Qwen2.5 VL 7B released?
Qwen2.5 VL 7B's default version was released on Jan 26, 2025.
How much does Qwen2.5 VL 7B cost?
Qwen2.5 VL 7B's official API price is $0.35 per million input tokens and $1.05 per million output tokens via Alibaba.
Who created Qwen2.5 VL 7B?
Qwen2.5 VL 7B was created by Alibaba Cloud / Qwen Team.
What is the context window for Qwen2.5 VL 7B?
A context window is not available for the default version.
Is Qwen2.5 VL 7B open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Qwen2.5 VL 7B?
3 provider offerings are linked to the default version.
What models should I compare Qwen2.5 VL 7B with?
Nearby ranked alternatives include Phi 4 Mini Reasoning, Nova Lite, Llama 3.1 70B.