Gemini 2.5 Flash Preview 05-20
| Released | 2025-05-20 |
|---|---|
| Price per 1M | $0.14 in · $3.33 out |
| Overall score | 46.2 · rank 149 |
| Context window | 1,048,576 tokens |
| Provider | |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 46.2 | 149 | 8 |
| Overall: Coding | 44.2 | 162 | 4 |
| Overall: Reasoning | 47.4 | 138 | 5 |
| Overall: Agentic work | 45.0 | 158 | 4 |
| Overall: Frontend | 46.1 | 127 | 2 |
| Coding | 43.6 | 165 | 1 |
| Reasoning | 47.2 | 135 | 3 |
| Math | 48.7 | 124 | 1 |
| Knowledge | 45.3 | 114 | 1 |
| General | 45.4 | 140 | 1 |
| Multimodal | 53.8 | 58 | 1 |
| Value vs price | −1.4 | 138 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 45.1 ±2.3 | 8 |
| Thinking | 45.1 ±4.0 | 3 |
Evidence
Coding 43.6 ±1.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Aider Polyglot | 55.1% | Epoch AI | 2026-11-11 |
Reasoning 47.2 ±3.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 2.5% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 | 33.3% | Epoch AI | 2026-11-11 |
| EnigmaEval | 2.7% | Epoch AI | 2026-11-11 |
Math 48.7 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 70.8% | Epoch AI | 2025-05-20 |
Knowledge 45.3 ±1.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Humanity's Last Exam | 11.0% | Epoch AI | 2026-11-11 |
General 45.4 ±5.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| EQ-Bench Longform Creative Writing | 48.6% | EQ-Bench | 2026-11-11 |
Multimodal 53.8 ±16.1
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-1 | 33.3% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 1.7% | Epoch AI | A stronger configuration result was used |
| Aider Polyglot | 44.0% | Epoch AI | A stronger configuration result was used |
Names matched to this model
gemini-2-5-flash-preview-05-20