Multimodal RewardBench 2
Version 2
Reward models compare two outputs containing text and images. Accuracy measures agreement with expert preferences across four tasks.
The evaluation contains 4,000 labeled pairs.
- Released
- December 2025
- Kind
- General visual questions
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Preference
- Language
- English
- Interaction
- Single turn
- Items from
- Another benchmark, Generated, Written by experts
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- Multimodal RewardBench
- In views
- Vision
- Strongest model tested
-
Gemini 3 Pro #56 on the capability index
- Last model added
- November 2025
- Scores refreshed
- November 20th, 2026
Gemini 3 Pro leads with 76.3% among models released by November 2025 Gemini 3 Pro leads with 76.3% among models released by November 2025
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in November 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1
Gemini 3 Pro 76.3%
- 2 GPT-5 72.2%
- 3
Gemini 2.5 Pro Jun 2025 70.9%
- 4 Qwen3-VL-32B 64.6%
- 5
Gemini 2.5 Flash Sep 2025 64.1%
- 6 GPT-4.1 63.5%
- 7
Qwen3-VL-235B-A22B 62.9%
- 8 GPT-4o Mar 2025 59.7%
- 9 Qwen3-VL-8B 59.3%
- 10
Qwen2.5-VL-72B 59%
- 11 Qwen3-VL-30BA3B 58.5%
- 12
Gemma 3 27B 57.3%
- 13
Gemma 3 12B 55.3%
- 14
Qwen2.5-VL-7B 50.9%
- 15
Gemma 3 4B 50.7%
Source: Official Multimodal RewardBench 2 leaderboard
- 1
Gemini 3 Pro 76.3%
- 2 GPT-5 72.2%
- 3
Gemini 2.5 Pro Jun 2025 70.9%
- 4
Gemini 2.5 Flash Sep 2025 64.1%
- 5 GPT-4.1 63.5%
- 6
Qwen3-VL-235B-A22B 62.9%
- 7 GPT-4o Mar 2025 59.7%
- 8
Qwen2.5-VL-72B 59%
- 9
Gemma 3 27B 57.3%
- 10
Gemma 3 12B 55.3%
- 11
Qwen2.5-VL-7B 50.9%
- 12
Gemma 3 4B 50.7%
Source: Official Multimodal RewardBench 2 leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 76.3% | Nov 2025 | Official leaderboard | |
| 2 | GPT-5 | 72.2% | Aug 2025 | Official leaderboard | |
| 3 | | 70.9% | Jun 2025 | Official leaderboard | |
| 4 | Qwen3-VL-32B open weights | 64.6% | Official leaderboard | ||
| 5 | | 64.1% | Sep 2025 | Official leaderboard | |
| 6 | GPT-4.1 | 63.5% | Apr 2025 | Official leaderboard | |
| 7 | | 62.9% | Sep 2025 | Official leaderboard | |
| 8 | GPT-4o Mar 2025 | 59.7% | Mar 2025 | Official leaderboard | |
| 9 | Qwen3-VL-8B open weights | 59.3% | Official leaderboard | ||
| 10 | | 59% | Jan 2025 | Official leaderboard | |
| 11 | Qwen3-VL-30BA3B open weights | 58.5% | Official leaderboard | ||
| 12 | | 57.3% | Mar 2025 | Official leaderboard | |
| 13 | | 55.3% | Mar 2025 | Official leaderboard | |
| 14 | | 50.9% | Jan 2025 | Official leaderboard | |
| 15 | | 50.7% | Mar 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 76.3% | Nov 2025 | Official leaderboard | |
| 2 | GPT-5 | 72.2% | Aug 2025 | Official leaderboard | |
| 3 | | 70.9% | Jun 2025 | Official leaderboard | |
| 4 | | 64.1% | Sep 2025 | Official leaderboard | |
| 5 | GPT-4.1 | 63.5% | Apr 2025 | Official leaderboard | |
| 6 | | 62.9% | Sep 2025 | Official leaderboard | |
| 7 | GPT-4o Mar 2025 | 59.7% | Mar 2025 | Official leaderboard | |
| 8 | | 59% | Jan 2025 | Official leaderboard | |
| 9 | | 57.3% | Mar 2025 | Official leaderboard | |
| 10 | | 55.3% | Mar 2025 | Official leaderboard | |
| 11 | | 50.9% | Jan 2025 | Official leaderboard | |
| 12 | | 50.7% | Mar 2025 | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Text-to-Image, Image Editing and Interleaved. Gemini 3 Pro leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
Text-to-Image
12 models we track have a score here.
- 1
Gemini 3 Pro 74.4%
- 2
Gemini 2.5 Pro Jun 2025 ↑1 70.5%
- 3 GPT-5 ↓1 70.5%
- 4 GPT-4.1 ↑1 65.8%
- 5
Gemini 2.5 Flash Sep 2025 ↓1 63.1%
- 6
Qwen3-VL-235B-A22B 62%
- 7 GPT-4o Mar 2025 60.3%
- 8
Qwen2.5-VL-72B 59.1%
- 9
Gemma 3 27B 58.3%
- 10
Gemma 3 12B 56%
Image Editing
12 models we track have a score here.
- 1
Gemini 3 Pro 74.9%
- 2 GPT-5 73.8%
- 3
Gemini 2.5 Pro Jun 2025 71.3%
- 4 GPT-4.1 ↑1 68.2%
- 5
Gemini 2.5 Flash Sep 2025 ↓1 66.5%
- 6 GPT-4o Mar 2025 ↑1 65%
- 7
Qwen3-VL-235B-A22B ↓1 64.8%
- 8
Qwen2.5-VL-72B 64.6%
- 9
Gemma 3 27B 60.2%
- 10
Gemma 3 12B 58%
Interleaved
12 models we track have a score here.
- 1
Gemini 3 Pro 76.4%
- 2
Gemini 2.5 Pro Jun 2025 ↑1 75.1%
- 3 GPT-5 ↓1 74.4%
- 4
Gemini 2.5 Flash Sep 2025 69.4%
- 5
Qwen3-VL-235B-A22B ↑1 69%
- 6 GPT-4.1 ↓1 67%
- 7
Qwen2.5-VL-72B ↑1 62.3%
- 8 GPT-4o Mar 2025 ↓1 61.5%
- 9
Gemma 3 27B 61.1%
- 10
Gemma 3 12B 58%
Reasoning
12 models we track have a score here.
- 1
Gemini 3 Pro 79.5%
- 2 GPT-5 70.2%
- 3
Gemini 2.5 Pro Jun 2025 66.6%
- 4
Gemini 2.5 Flash Sep 2025 57.5%
- 5
Qwen3-VL-235B-A22B ↑1 55.9%
- 6 GPT-4.1 ↓1 53%
- 7 GPT-4o Mar 2025 51.9%
- 8
Qwen2.5-VL-72B 50%
- 9
Gemma 3 27B 49.4%
- 10
Gemma 3 12B 49.3%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Multimodal RewardBench 2 leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Multimodal RewardBench 2 leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
Versions
We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.
Source: Official leaderboards
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy.
Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy.
Qwen 3.8 Max 85.9%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 12 months ago, in November 2025, for Gemini 3 Pro. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the official leaderboard (15 entries).
- Pass Models scored 12 models we track, plus 3 other entries on its leaderboard.
- Warning Newest score The newest came 12 months ago, in November 2025, for Gemini 3 Pro.
- Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
- Pass Headroom The best is 76.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 12 rows for models we track and 3 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.