Benchmark Almanac

LiveOCRVQA

Gemini 2.5 Pro Jun 2025 80.5 leads since June 2025
Not in model cards

Models read stylized text in images and answer questions about it.

The publisher reports model results.

Made by
BAAI FlagEval
Kind
Documents, charts and text in images
Level
Everyday
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Images → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
In views
Vision
Strongest model tested
Gemini 2.5 Pro Jun 2025 #97 on the capability index
Last model added
September 2025
Scores refreshed
November 20th, 2026
21
Entries on its own leaderboard, 18 models we track

Gemini 2.5 Pro leads with 80.5 among models released by September 2025 Gemini 2.5 Pro leads with 80.5 among models released by September 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 September 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 Gemini 2.5 Pro Jun 2025 80.5
  2. 2 Gemini 2.5 Flash Sep 2025 75.3
  3. 3 Qwen2-VL-7B-Instruct 72.2
  4. 4 GPT-4o-2411 69.1
  5. 5 GPT-4o mini 67.5
  6. 6 Qwen2.5-VL-72B 67.3
  7. 7 GPT-4o-2408 67.0
  8. 8 Qwen2.5-VL-7B 66.0
  9. 9 Qwen2-VL-2B-Instruct 64.7
  10. 10 Qwen2-VL-72B 63.6
  11. 11 InternVL3-78B 63.6
  12. 12 Claude 3.7 Sonnet 62.6
  13. 13 Claude 3.5 Sonnet 59.7
  14. 14 Pixtral Large 59.0
  15. 15 InternVL2_5-78B 50.6
  16. 16 InternVL3-8B 49.6
  17. 17 LLaVA-OneVision 7B 45.7
  18. 18 InternVL2.5-8B 39.2
  19. 19 Phi-3.5-vision-instruct 37.4
  20. 20 Idefics3-8B-Llama3 28.1
  21. 21 Phi-4-multimodal 17.9

Source: Official LiveOCRVQA leaderboard

#EntryScoreSettingDateSource
1 Gemini 2.5 Pro Jun 2025 80.5 Jun 2025 Official leaderboard
2 Gemini 2.5 Flash Sep 2025 75.3 Sep 2025 Official leaderboard
3 Qwen2-VL-7B-Instruct 72.2 Aug 2024 Official leaderboard
4 GPT-4o-2411 69.1 Official leaderboard
5 GPT-4o mini 67.5 Jul 2024 Official leaderboard
6 Qwen2.5-VL-72B 67.3 Jan 2025 Official leaderboard
7 GPT-4o-2408 67.0 Official leaderboard
8 Qwen2.5-VL-7B 66.0 Jan 2025 Official leaderboard
9 Qwen2-VL-2B-Instruct 64.7 Aug 2024 Official leaderboard
10 Qwen2-VL-72B 63.6 Official leaderboard
11 InternVL3-78B 63.6 Apr 2025 Official leaderboard
12 Claude 3.7 Sonnet 62.6 Feb 2025 Official leaderboard
13 Claude 3.5 Sonnet 59.7 Jun 2024 Official leaderboard
14 Pixtral Large 59.0 Nov 2024 Official leaderboard
15 InternVL2_5-78B 50.6 Dec 2024 Official leaderboard
16 InternVL3-8B 49.6 Apr 2025 Official leaderboard
17 LLaVA-OneVision 7B 45.7 Aug 2024 Official leaderboard
18 InternVL2.5-8B 39.2 Dec 2024 Official leaderboard
19 Phi-3.5-vision-instruct 37.4 Aug 2024 Official leaderboard
20 Idefics3-8B-Llama3 28.1 Aug 2024 Official leaderboard
21 Phi-4-multimodal 17.9 Feb 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 2.5 Pro Jun 2025 80.5 Jun 2025 Official leaderboard
2 Gemini 2.5 Flash Sep 2025 75.3 Sep 2025 Official leaderboard
3 Qwen2-VL-7B-Instruct 72.2 Aug 2024 Official leaderboard
4 GPT-4o mini 67.5 Jul 2024 Official leaderboard
5 Qwen2.5-VL-72B 67.3 Jan 2025 Official leaderboard
6 Qwen2.5-VL-7B 66.0 Jan 2025 Official leaderboard
7 Qwen2-VL-2B-Instruct 64.7 Aug 2024 Official leaderboard
8 InternVL3-78B 63.6 Apr 2025 Official leaderboard
9 Claude 3.7 Sonnet 62.6 Feb 2025 Official leaderboard
10 Claude 3.5 Sonnet 59.7 Jun 2024 Official leaderboard
11 Pixtral Large 59.0 Nov 2024 Official leaderboard
12 InternVL2_5-78B 50.6 Dec 2024 Official leaderboard
13 InternVL3-8B 49.6 Apr 2025 Official leaderboard
14 LLaVA-OneVision 7B 45.7 Aug 2024 Official leaderboard
15 InternVL2.5-8B 39.2 Dec 2024 Official leaderboard
16 Phi-3.5-vision-instruct 37.4 Aug 2024 Official leaderboard
17 Idefics3-8B-Llama3 28.1 Aug 2024 Official leaderboard
18 Phi-4-multimodal 17.9 Feb 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 Album, Book and Game. Gemini 2.5 Pro leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

Album

18 models we track have a score here.

  1. 1 Gemini 2.5 Pro Jun 2025 68.3
  2. 2 Gemini 2.5 Flash Sep 2025 66.2
  3. 3 Qwen2-VL-7B-Instruct 64.1
  4. 4 GPT-4o mini 54.9
  5. 5 InternVL3-78B ↑3 54.2
  6. 6 Qwen2.5-VL-72B ↓1 53.5
  7. 7 Qwen2-VL-2B-Instruct 52.8
  8. 8 Qwen2.5-VL-7B ↓2 51.4
  9. 9 LLaVA-OneVision 7B ↑5 45.8
  10. 10 InternVL2_5-78B ↑2 43.7

4 models have held number one on the official leaderboard since 2023

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

Anthropic OpenAI Other labs Google DeepMind
Own leaderboard Not tested since Sep 2025

Source: Official LiveOCRVQA 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
0.00 20.0 40.0 60.0 80.0 100.0 2024 2025 0.00 20.0 40.0 60.0 80.0 100.0 2024 2025

Source: Official LiveOCRVQA 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

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (21 entries).
  • Pass Models scored 18 models we track, plus 3 other entries on its leaderboard.
  • Warning Newest score The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 2.5 Pro (#97).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on flageval-baai.github.io: 18 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.

How we check