Skip to main content
BenchLM

RealWorldQA

We show this table for reference; we do not rank on it.

Data verified 34 confirmed releases in the last 30 daysFollow model changes

A grounded visual QA benchmark focused on answering practical questions about real-world images and scenes.

Benchmark score on RealWorldQA — September 27, 2026

We compile the RealWorldQA rows from provider self-reports. Qwen3.8-Flash-Next leads the table at 88.5%, followed by Qwen3.8 Max (88.0%) and Qwen3.8-Omni-Flash (87.7%). We do not use these results to rank models overall.

1Open weights

Qwen3.8-Flash-Next

Alibaba

88.5%
Overall 60.72Context 262K
2Open weights

Qwen3.8 Max

Alibaba

88.0%
Overall 71.8Context 1M
3Proprietary

Qwen3.8-Omni-Flash

Alibaba

87.7%
Context 1M

12 modelsMultimodal & GroundedCurrentDisplay onlyUpdated September 27, 2026

Benchmark score table (12 models)

Score
1
Qwen3.8-Flash-NextAlibaba · Open weight
88.5%
2
Qwen3.8 MaxAlibaba · Open weight
88.0%
3
Qwen3.8-Omni-FlashAlibaba · Closed
87.7%
4
Qwen3.7 PlusAlibaba · Closed
86.9%
5
Qwen3.8-27BAlibaba · Open weight
85.9%
6
Qwen3.6-35B-A3BAlibaba · Open weight
85.3%
7
Qwen3.6-27BAlibaba · Open weight
84.1%
8
Ternary Bonsai 2 27BPrism ML · Open weight
80.1%
9
LFM2.5-VL-3BLiquidAI · Open weight
73.1%
10
ZAYA1-VL-8BZyphra · Open weight
65.0%
11
North Micro Vision InstructCohere · Open weight
62.2%
12
LFM2.5-VL-450MLiquidAI · Open weight
58.4%

Among the reported RealWorldQA rows, Qwen3.8-Flash-Next is first at 88.5%. The third row is 0.8 points behind. The broader top-10 range is 23.5 points, so the table still separates the published systems.

12 models have been evaluated on RealWorldQA. The benchmark falls in the Multimodal & Grounded category. RealWorldQA is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About RealWorldQA

Year

2026

Tasks

Real-world visual question answering

Format

Image-grounded QA

Difficulty

General visual reasoning

RealWorldQA is useful because it emphasizes practical perception and grounded answering on realistic images rather than synthetic or purely academic multimodal tasks.

Qwen3.6 launch benchmarks

Freshness and provenance

Version

RealWorldQA 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

What does RealWorldQA measure?

A grounded visual QA benchmark focused on answering practical questions about real-world images and scenes.

Which model scores highest on RealWorldQA?

Qwen3.8-Flash-Next by Alibaba currently leads with a score of 88.5% on RealWorldQA.

How many models are evaluated on RealWorldQA?

12 AI models have been evaluated on RealWorldQA on BenchLM.

Compare top models on RealWorldQA

Qwen3.8-Flash-Next vs Qwen3.8 MaxQwen3.8 Max vs Qwen3.8-Omni-FlashQwen3.8-Omni-Flash vs Qwen3.7 PlusQwen3.7 Plus vs Qwen3.8-27B
Last updated: September 27, 2026 · BenchLM version RealWorldQA 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.