Gemini-3-Flash
Google1000K context2026-02-01
Overall Rank
#4
of 29 models
Overall Score
82.0
avg across benchmarks
Best Task
Text Extraction
92.3
Weakest Task
Visual QA
63.5
Benchmark Performance
OlmOCR Benchv1.0
11/29
| Overall | ArXiv Math | H&F | Long/Tiny | Multi-Col | Old Scans | Scans Math | Tables |
|---|---|---|---|---|---|---|---|
| 75.3 | 80.1 | 27.4 | 90.3 | 75.3 | 45.8 | 73.6 | 64.6 |
OmniDocBenchv1.5
1/29
| Overall | Text Edit↓ | CDM↑ | TEDS↑ | TEDS-S↑ | Read Order↓ |
|---|---|---|---|---|---|
| 90.1 | 0.077 | 90.2 | 87.7 | 92.6 | 0.081 |
IDP Core Benchv1.0
6/29
| Overall | KIE | OCR | Table | VQA |
|---|---|---|---|---|
| 80.5 | 91.1 | 81.7 | 85.6 | 63.5 |
Capability Profile
Strength Analysis
Auto-generated from benchmark scores
Strengths
- Text Extraction92.3
- Key Information Extraction91.1
Weaknesses
- Visual QA63.5
- Table Understanding79.3