MIA-Bench
Qwen3-VL-235B-A22B 92.7% leads since Sep 2025 Dropped Models answer 400 image prompts that stack several strict instructions on format, length and content. An LLM judge scores how well each answer follows every instruction.
ICLR 2025 paper. Reported in Qwen-VL reports under alignment.
In the authors’ words
We introduce MIA-Bench, a new benchmark designed to evaluate multimodal large language models (MLLMs) on their ability to strictly adhere to complex instructions. Our benchmark comprises a diverse set of 400 image-prompt pairs, each crafted to challenge the models' compliance with layered instructions in generating accurate responses that satisfy specific requested patterns. Evaluation results from a wide array of state-of-the-art MLLMs reveal significant variations in performance, highlighting areas for improvement in instruction fidelity. Additionally, we create extra training data and explore supervised fine-tuning to enhance the models' ability to strictly follow instructions without compromising performance on other tasks. We hope this benchmark not only serves as a tool for measuring MLLM adherence to instructions, but also guides future developments in MLLM training methods.
- Level
- Everyday Most adults, without training
- Scored
- Graded by a model A language model scores each response, usually against a rubric.
- Answers
- Open-ended Free-form text, documents, designs or media judged as a whole.
Qwen3-VL-235B-A22B leads with 92.7%
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Qwen3-VL-235B-A22B | 92.7% | Sep 2025 | Model card | |
| 2 |  STEP3-VL-10B | 92% | Jan 2026 | Model card |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
The best score passed 85% in Sep 2025
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Labs dropped it from launch posts once the best score passed 85%
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 23 Sep 2025 | Qwen3-VL-235B-A22B | 92.7 | MIABench |
2 system cards report it, from 2 labs
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 14 Jan 2026 | STEP3-VL-10B technical report | 92 | MIA-Bench, SeRe (Table 1) | Capabilities |
| 26 Nov 2025 | Qwen3-VL-235B-A22B technical report | 92.7 | MIA-Bench (Table 2) | Capabilities |
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. GPT-5.1 85.4%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. Kimi K3.1 Pro 84.1%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
- ChartQA In use Models answer human-written and generated questions about chart images that need visual and arithmetic reasoning. Scored by relaxed accuracy with a 5 percent numeric tolerance.
Command A Vision 90.9% - DocVQA In use Models answer 50,000 questions about 12,000 scanned document images by reading their text and layout. Scoring is ANLS, a string similarity to reference answers.
Seed1.5-VL 96.9%