Benchmark Almanac

RBench-M

GPT-5 69.9% leads since August 2025
Not in model cards

Models answer 665 graduate-level multimodal reasoning questions in English and Chinese. Top-1 accuracy compares selected answers with the answer key.

This is the multimodal track of the multidisciplinary RBench. It is distinct from the existing rbench record about video generation. English and Chinese results are separate language tracks.

Released
May 2025
Kind
General visual questions
Level
Expert
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
English, Chinese
In and out
Text, Images → Text
Interaction
Single turn
Items from
Another benchmark
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-5 #71 on the capability index
Last model added
October 2025
Scores refreshed
November 20th, 2026

In the authors’ words

Reasoning stands as a cornerstone of intelligence, enabling the synthesis of existing knowledge to solve complex problems. Despite remarkable progress, existing reasoning benchmarks often fail to rigorously evaluate the nuanced reasoning capabilities required for complex, real-world problemsolving, particularly in multi-disciplinary and multimodal contexts. In this paper, we introduce a graduate-level, multi-disciplinary, EnglishChinese benchmark, dubbed as Reasoning Bench (R-Bench), for assessing the reasoning capability of both language and multimodal models. RBench spans 1,094 questions across 108 subjects for language model evaluation and 665 questions across 83 subjects for multimodal model testing in both English and Chinese. These questions are meticulously curated to ensure rigorous difficulty calibration, subject balance, and crosslinguistic alignment, enabling the assessment to be an Olympiad-level multi-disciplinary benchmark. We evaluate widely used models, including OpenAI o1, GPT-4o, DeepSeek-R1, etc. Experimental results indicate that advanced models perform poorly on complex reasoning, especially multimodal reasoning. Even the top-performing model OpenAI o1 achieves only 53.2% accuracy on our multimodal evaluation. Data and code are made publicly available at here.

Abstract of R-Bench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation (arXiv:2505.02018, CC0)
15
Entries on its own leaderboard, 13 models we track

GPT-5 leads with 69.9% among models released by October 2025 GPT-5 leads with 69.9% among models released by October 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 October 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-5 69.9%
  2. 2 Doubao-Seed-1.6 63.3%
  3. 3 o1 53.1%
  4. 4 GLM-4.5V 41%
  5. 5 Doubao-1.5-pro 40.2%
  6. 6 Claude 3.5 Sonnet (October 2024) 39%
  7. 7 JT-VL-Chat-Thinking-20251015 37.8%
  8. 8 Qwen3-VL-30B-A3B-Instruct 36.8%
  9. 9 Gemini 1.5 Pro Feb 2024 35.7%
  10. 10 GPT-4o Nov 2024 33.3%
  11. 11 Qwen2-VL-72B 25.4%
  12. 12 LLaVA-OneVision 7B 23.7%
  13. 13 DeepSeek-VL2 23.1%
  14. 14 Qwen2-VL-7B-Instruct 21%
  15. 15 Llama 3.2 11B 19.3%

Source: Official RBench-M leaderboard

#EntryScoreSettingDateSource
1 GPT-5 69.9% Aug 2025 Official leaderboard
2 Doubao-Seed-1.6 63.3% Jun 2025 Official leaderboard
3 o1 53.1% Dec 2024 Official leaderboard
4 GLM-4.5V 41% Aug 2025 Official leaderboard
5 Doubao-1.5-pro 40.2% Jan 2025 Official leaderboard
6 Claude 3.5 Sonnet (October 2024) 39% Oct 2024 Official leaderboard
7 JT-VL-Chat-Thinking-20251015 37.8% Official leaderboard
8 Qwen3-VL-30B-A3B-Instruct 36.8% Oct 2025 Official leaderboard
9 Gemini 1.5 Pro Feb 2024 35.7% Feb 2024 Official leaderboard
10 GPT-4o Nov 2024 33.3% Nov 2024 Official leaderboard
11 Qwen2-VL-72B 25.4% Official leaderboard
12 LLaVA-OneVision 7B 23.7% Aug 2024 Official leaderboard
13 DeepSeek-VL2 23.1% Dec 2024 Official leaderboard
14 Qwen2-VL-7B-Instruct 21% Aug 2024 Official leaderboard
15 Llama 3.2 11B 19.3% Sep 2024 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5 69.9% Aug 2025 Official leaderboard
2 Doubao-Seed-1.6 63.3% Jun 2025 Official leaderboard
3 o1 53.1% Dec 2024 Official leaderboard
4 GLM-4.5V 41% Aug 2025 Official leaderboard
5 Doubao-1.5-pro 40.2% Jan 2025 Official leaderboard
6 Claude 3.5 Sonnet (October 2024) 39% Oct 2024 Official leaderboard
7 Qwen3-VL-30B-A3B-Instruct 36.8% Oct 2025 Official leaderboard
8 Gemini 1.5 Pro Feb 2024 35.7% Feb 2024 Official leaderboard
9 GPT-4o Nov 2024 33.3% Nov 2024 Official leaderboard
10 LLaVA-OneVision 7B 23.7% Aug 2024 Official leaderboard
11 DeepSeek-VL2 23.1% Dec 2024 Official leaderboard
12 Qwen2-VL-7B-Instruct 21% Aug 2024 Official leaderboard
13 Llama 3.2 11B 19.3% Sep 2024 Official leaderboard

Rankings by part

We read 2 parts from the benchmark’s own leaderboard: RBench-M (English) and RBench-M (Chinese). Each ranks models on one part of the test. GPT-5 leads both. Arrows show how far a model moved from its overall rank among the models we track.

RBench-M (English)

13 models we track have a score here.

  1. 1 GPT-5 68.4%
  2. 2 Doubao-Seed-1.6 62.4%
  3. 3 o1 53.2%
  4. 4 GLM-4.5V 40.8%
  5. 5 Claude 3.5 Sonnet (October 2024) ↑1 39.7%
  6. 6 Doubao-1.5-pro ↓1 37.9%
  7. 7 Qwen3-VL-30B-A3B-Instruct 36.5%
  8. 8 Gemini 1.5 Pro Feb 2024 35.9%
  9. 9 GPT-4o Nov 2024 33.4%
  10. 10 LLaVA-OneVision 7B 23.8%

3 models have held number one on the official leaderboard since its release in May 2025

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI ByteDance Seed
Own leaderboard Not tested since Oct 2025
GPT-5 69.9%

Source: Official RBench-M leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) GPT-5 Doubao-Seed-1.6 o1 Claude 3.5 Sonnet (October 2024) Gemini 1.5 Pro 0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) GPT-5 Doubao-Seed-1.6

Source: Official RBench-M 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.

More in Vision

All 636 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 13 months ago, in October 2025, for Qwen3-VL-30B-A3B-Instruct. 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 13 models we track, plus 2 other entries on its leaderboard.
  • Warning Newest score The newest came 13 months ago, in October 2025, for Qwen3-VL-30B-A3B-Instruct.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
  • Pass Headroom The best is 69.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on evalmodels.github.io: 13 rows for models we track and 2 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.

How we check