Gemma 3 27B IT
| Released | 2025-03-12 |
|---|---|
| Price per 1M | $0.12 in · $0.30 out |
| Overall score | 34.1 · rank 224 |
| Context window | 131,072 tokens |
| Provider | |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 34.1 | 224 | 13 |
| Overall: Coding | 27.7 | 230 | 6 |
| Overall: Reasoning | 39.0 | 199 | 5 |
| Overall: Agentic work | 32.2 | 226 | 6 |
| Overall: Frontend | 28.2 | 219 | 4 |
| Coding | 23.4 | 209 | 2 |
| Agents | 34.4 | 173 | 1 |
| Reasoning | 38.1 | 205 | 3 |
| Math | 40.1 | 161 | 2 |
| General | 43.7 | 144 | 3 |
| Multimodal | 42.4 | 100 | 2 |
| Value vs price | −8.3 | 171 |
Evidence
Coding 23.4 ±1.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Aider Polyglot | 4.9% | Epoch AI | 2026-11-11 |
| SciCode | 23.3% | Epoch AI | 2026-11-11 |
Agents 34.4 ±3.6
Reasoning 38.1 ±4.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond | 47.7% | Epoch AI | 2026-08-27 |
| CritPt | 0.0% | Epoch AI | 2026-11-11 |
| Kagi Reasoning Benchmark | 35.1% | Kagi | 2026-11-11 |
Math 40.1 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MATH Level 5 | 74.0% | Epoch AI | 2025-03-13 |
| OTIS Mock AIME 2024-2025 | 22.5% | Epoch AI | 2026-08-27 |
General 43.7 ±3.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,365 | LMArena | 2026-11-09 |
| EQ-Bench Creative Writing v3 | 1,266 | EQ-Bench | 2026-11-11 |
| EQ-Bench Longform Creative Writing | 41.9% | EQ-Bench | 2026-11-11 |
Multimodal 42.4 ±4.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Vision | 1,160 | LMArena | 2026-11-09 |
| GeoBench | 52.0% | Epoch AI | 2026-11-11 |
Names matched to this model
gemma-3-27b, gemma-3-27b-it