Qwen2-VL-72B-Instruct
Released
2025-01-01
Last refreshed
2026-10-07
Status
Researched 180d ago
Open sourceCommercial use: permittedMultimodalVision
Alibaba releases · 51 in the last 12 monthsChangelog →
Specifications
- Family
- Qwen2-VL
- Released
- 2025-01-01
- Context
- 32K
- Parameters
- 72B
- Architecture
- Decoder Only
- Knowledge cutoff
- 2023-06
- Specialization
- multimodal
- Openness
- Open source
- License
- Apache 2.0OSI-approvedPermitted
- Weights
- Unknown
- Code
- Unknown
- Training
- Pretrained
Created by
Pricing
Output / 1M
$0.900
Input / 1M
$0.900
Cheapest of 1 route · Fireworks AI
Providers(1)
Created by
Pricing
Output / 1M
$0.900
Input / 1M
$0.900
Cheapest of 1 route · Fireworks AI
Providers(1)
About
Qwen2-VL-72B-Instruct is Alibaba's Qwen2-VL model focused on multimodal input across text, image, and beyond. It offers a 32K-token context window.
Top use-case fit
Vision
Q/$ BProviders(1)
Compare API pricing across 1 tracked provider route for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Route |
|---|---|---|---|
| Fireworks AI | $0.900 | $0.900 | Serverless |
Available via routers & gateways(1)
Capabilities
VisionMultimodal
Benchmark peer barsfor Vision
Benchmark scores(2)
Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| 64.5 | —Observed 2026-04-15 | — | ||
| 59.3 | standard 4-option (original paper harness)Observed 2024-09-04 | — |
Migration checks
No linked migration route is available for this model yet.