Alibaba Cloud / Qwen Team model product
Qwen2 VL 72B
Qwen2 VL 72B is an instruction-tuned large multimodal model for image and video input, visual understanding, and step-by-step reasoning.
Updated Sep 8, 2026. Default version: Qwen2-VL-72B-Instruct
LLMBoard Score24.4Qwen2-VL-72B-Instruct
Coverage0%12 benchmark families
Context windowN/ATokens
Official input priceN/AOfficial price unavailable
Qwen2 VL 72B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Qwen2 VL 72B Benchmark Results
Benchmark scores for Qwen2-VL-72B-Instruct.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkEgoSchema | Score77.90% | Rank01 | Participants9 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMMVetGPT4Turbo | Score74.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMTVQA | Score30.90% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTextVQA | Score85.50% | Rank01 | Participants16 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVCR_en_easy | Score91.93% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkDocVQAtest | Score96.50% | Rank03 | Participants11 | Percentile80.00% | EvidenceC | Evaluated |
| BenchmarkMMBench | Score86.50% | Rank04 | Participants9 | Percentile62.50% | EvidenceC | Evaluated |
| BenchmarkMMMUval | Score64.50% | Rank04 | Participants4 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkChartQA | Score88.30% | Rank06 | Participants26 | Percentile80.00% | EvidenceC | Evaluated |
| BenchmarkMVBench | Score73.60% | Rank07 | Participants18 | Percentile64.71% | EvidenceC | Evaluated |
| BenchmarkInfoVQAtest | Score84.50% | Rank08 | Participants12 | Percentile36.36% | EvidenceC | Evaluated |
| BenchmarkOCRBench | Score87.70% | Rank12 | Participants24 | Percentile52.17% | EvidenceC | Evaluated |
| BenchmarkRealWorldQA | Score77.80% | Rank18 | Participants31 | Percentile43.33% | EvidenceC | Evaluated |
| BenchmarkMathVista-Mini | Score70.50% | Rank20 | Participants25 | Percentile20.83% | EvidenceC | Evaluated |
| BenchmarkMMMU-Pro | Score46.20% | Rank62 | Participants69 | Percentile10.29% | EvidenceC | Evaluated |
| BenchmarkLM Arena Vision | Score1,046.93 rating | Rank96 | Participants107 | Percentile10.38% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,085.24 rating | Rank96 | Participants107 | Percentile10.38% | EvidenceA | Evaluated |
Qwen2 VL 72B Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision style control | Categorychinese | Rank81 | Rating / score1,108.23 | Votes291 | ObservationsN/A | Result date |
| Arenavision | Categorychinese | Rank82 | Rating / score1,070.69 | Votes291 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank96 | Rating / score1,046.93 | Votes5,937 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank96 | Rating / score1,085.24 | Votes5,937 | ObservationsN/A | Result date |
| Arenavision | Categoryenglish | Rank97 | Rating / score1,053.47 | Votes3,230 | ObservationsN/A | Result date |
| Arenavision style control | Categoryenglish | Rank97 | Rating / score1,084.31 | Votes3,230 | ObservationsN/A | Result date |
Qwen2 VL 72B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Qwen2 VL 72B Runtime Performance
Provider-specific output speed and catalog latency for Qwen2-VL-72B-Instruct. Runtime does not affect the capability score.
No runtime data
No 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 VL 72B Specifications
Technical details for the model's default version.
- Version
- Qwen2-VL-72B-Instruct
- Released
- Aug 29, 2024
- Knowledge cutoff
- Jun 30, 2023
- Parameters
- 73.4B
- Context window
- N/A
- Max output
- N/A
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- tongyi-qianwen
Qwen2 VL 72B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionQwen2-VL-72B-Instruct | Released | LLMBoard24.40 | Parameters73.4B | ContextN/A | Max outputN/A | Open weightsYes | Licensetongyi-qianwen |
Models similar to Qwen2 VL 72B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AC
QwQ 32B
Alibaba Cloud / Qwen Team
AC
Qwen3 VL 8B
Alibaba Cloud / Qwen Team
AC
Qwen3 VL 4B Thinking
Alibaba Cloud / Qwen Team
AC
Qwen2.5 VL 72B
Alibaba Cloud / Qwen Team
AC
Qwen3 VL 8B Thinking
Alibaba Cloud / Qwen Team
AC
Qwen3 VL 30B A3B
Alibaba Cloud / Qwen Team
What is Qwen2 VL 72B?
Key information about Qwen2 VL 72B and its available data.
Qwen2 VL 72B is a large multimodal language model from Alibaba Cloud / Qwen Team. It supports image and video input, dynamic resolution handling, M-ROPE, multilingual text recognition in images, complex problem solving, and agent-like interactions in video contexts.
Data as of 2026-09-08.
FAQ
Common questions about Qwen2 VL 72B.
When was Qwen2 VL 72B released?
Qwen2 VL 72B's default version was released on Aug 29, 2024.
How much does Qwen2 VL 72B cost?
No official standard PAYG price is currently available for Qwen2 VL 72B.
Who created Qwen2 VL 72B?
Qwen2 VL 72B was created by Alibaba Cloud / Qwen Team.
What is the context window for Qwen2 VL 72B?
A context window is not available for the default version.
Is Qwen2 VL 72B open weight?
Yes. The default version is marked as open weight under tongyi-qianwen.
How many API providers offer Qwen2 VL 72B?
No provider offering is currently linked to the default version.