Gemini 2.5 Flash (Apr 2025)
| Released | 2025-04-17 |
|---|---|
| Price per 1M | Not listed |
| Overall score | 44.8 · rank 158 |
| Provider | |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 44.8 | 158 | 9 |
| Overall: Coding | 42.1 | 178 | 4 |
| Overall: Reasoning | 46.8 | 141 | 4 |
| Overall: Agentic work | 42.9 | 173 | 4 |
| Overall: Frontend | 43.3 | 154 | 5 |
| Coding | 41.5 | 174 | 2 |
| Reasoning | 45.2 | 150 | 2 |
| Math | 49.2 | 122 | 1 |
| Knowledge | 46.3 | 110 | 1 |
| Multimodal | 48.4 | 78 | 3 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 44.8 ±2.6 | 9 |
| Thinking | 39.5 ±6.2 | 2 |
Evidence
Coding 41.5 ±1.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Aider Polyglot | 47.1% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 56.9% | Vals AI | 2026-09-01 |
Reasoning 45.2 ±3.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-1 | 32.3% | Epoch AI | 2026-11-11 |
| GPQA Diamond (Vals run) | 57.6% | Vals AI | 2026-09-01 |
Math 49.2 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | Epoch AI | 2025-04-22 |
Knowledge 46.3 ±1.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Humanity's Last Exam | 12.1% | Epoch AI | 2026-11-11 |
Multimodal 48.4 ±9.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| VPCT | 38.0% | Epoch AI | 2026-11-11 |
| MMMU-Pro (Vals run) | 71.9% | Vals AI | 2026-09-01 |
| GeoBench | 73.0% | Epoch AI | 2026-11-11 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| LiveCodeBench (Vals run) | 46.9% | Vals AI | A stronger configuration result was used |
| MMMU-Pro (Vals run) | 68.0% | Vals AI | A stronger configuration result was used |
Names matched to this model
gemini-2-5-flash-apr-2025, gemini-2-5-flash-preview-04-17, gemini-2-5-flash-preview-04-17-16k-thinking, gemini-2-5-flash-preview-04-17-24k-thinking