Skip to main content
BenchLM

AI2D test split (AI2D_TEST)

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 diagram understanding benchmark focused on scientific and educational visual question answering.

Benchmark score on AI2D_TEST — September 27, 2026

We compile the AI2D_TEST rows from provider self-reports. Qwen3.6-35B-A3B leads the table at 92.7%, followed by Nemotron 3 Nano Omni 30B A3B (88.5%) and ZAYA1-VL-8B (87.5%). We do not use these results to rank models overall.

1Open weights

Qwen3.6-35B-A3B

Alibaba

92.7%
Overall 41.88Context 262K
2Open weights

Nemotron 3 Nano Omni 30B A3B

NVIDIA

88.5%
Overall 30.84Context 256K
3Open weights

ZAYA1-VL-8B

Zyphra

87.5%
Context 131K

3 modelsMultimodal & GroundedCurrentDisplay onlyUpdated September 27, 2026

Benchmark score table (3 models)

Score
1
Qwen3.6-35B-A3BAlibaba · Open weight
92.7%
2
Nemotron 3 Nano Omni 30B A3BNVIDIA · Open weight
88.5%
3
ZAYA1-VL-8BZyphra · Open weight
87.5%

Among the reported AI2D_TEST rows, Qwen3.6-35B-A3B is first at 92.7%. The third row is 5.2 points behind. The broader top-10 range is 5.2 points, so many of the published results sit in a relatively narrow band.

3 models have been evaluated on AI2D_TEST. The benchmark falls in the Multimodal & Grounded category. AI2D_TEST is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About AI2D_TEST

Year

2026

Tasks

Diagram understanding

Format

Diagram-grounded QA

Difficulty

Structured visual reasoning

AI2D-style tasks matter because diagrams compress structure differently from photos or office documents. They test whether a model can parse arrows, labels, and spatial relations in technical illustrations.

Qwen3.6 launch benchmarks

Freshness and provenance

Version

AI2D_TEST 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 AI2D_TEST measure?

A diagram understanding benchmark focused on scientific and educational visual question answering.

Which model scores highest on AI2D_TEST?

Qwen3.6-35B-A3B by Alibaba currently leads with a score of 92.7% on AI2D_TEST.

How many models are evaluated on AI2D_TEST?

3 AI models have been evaluated on AI2D_TEST on BenchLM.

Compare top models on AI2D_TEST

Qwen3.6-35B-A3B vs Nemotron 3 Nano Omni 30B A3BNemotron 3 Nano Omni 30B A3B vs ZAYA1-VL-8B
Last updated: September 27, 2026 · BenchLM version AI2D_TEST 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.