Model comparison
Gemma 3 27B vs o3-mini
o3-mini is the stronger model overall, scoring 36.7 to 30.5 on the Noometry Index. Gemma 3 27B costs 19× less per token, which makes it the better buy when o3-mini's lead doesn't matter for your workload.
Last verified . 37 shared benchmarks.
GoogleSummary
- They share 37 benchmarks with published results for both. Gemma 3 27B scores higher in 2 categories and o3-mini in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where o3-mini leads 41.4 to 22.1.
- The biggest single-benchmark swing is Aider Polyglot: 4.9% for Gemma 3 27B and 60.4% for o3-mini.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- o3-mini accepts more context: 200K tokens versus 131K.
- Gemma 3 27B has downloadable open weights; the other is API-only.
Side by side
| Gemma 3 27B | o3-mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 30.5 | 36.7 |
| Released | 2025-03-11 | 2024-12-20 |
| Weights | Open | Proprietary |
| Context window | 131K | 200K |
| Max output | 8K | 100K |
| Input $ / M tokens | $0.08 | $1.10 |
| Output $ / M tokens | $0.16 | $4.40 |
| Results tracked | 43 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-mini leads
Gemma 3 27B: 22.1 (#329), o3-mini: 41.4 (#116)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| Aider Polyglot | 4.9% | 60.4% |
| SciCode | 21.2% | 39.8% |
| LiveBench Coding | 39.9% | 82.7% |
| LMArena Coding | 1322 | 1378 |
| GSO | — | 1.3% |
| WeirdML | — | 43.7% |
| CadEval | — | 54% |
Agentic & Tool Use o3-mini leads
Gemma 3 27B: 24.9 (#105), o3-mini: 29.3 (#79)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | — |
| Cybench | — | 22.5% |
Reasoning Too close to call
Gemma 3 27B: 16.3 (#297), o3-mini: 16.3 (#295)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| CritPt | 0% | 0.3% |
| Chess Puzzles | 0% | 17% |
| LiveBench Reasoning | 43.8% | 89.6% |
| LMArena Hard Prompts | 1340 | 1366 |
| DTBench | 52.5% | 68.8% |
| LiveBench Data Analysis | 51.5% | 70.6% |
| LMCA | 12.3% | 19% |
| Epoch Capabilities Index | 130.04 | 140.35 |
| LiveBench | 50% | 75.9% |
| ARC-AGI-2 | — | 3% |
| SimpleBench | — | 22.8% |
| Kagi LLM Benchmark | 40.4% | — |
| ARC-AGI-1 | — | 34.5% |
| Mystery Game Puzzles | — | 7% |
| ForecastBench | — | 59.6 |
Math o3-mini leads
Gemma 3 27B: 24.4 (#261), o3-mini: 27.6 (#236)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 76.9% |
| LiveBench Math | 55.4% | 77.3% |
| LMArena Math | 1312 | 1396 |
| MATH Level 5 | 74% | 96.5% |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| FrontierMath (Feb 2025 set) | — | 12.4% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge o3-mini leads
Gemma 3 27B: 25.4 (#256), o3-mini: 38.3 (#135)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| GPQA Diamond | 47.7% | 77% |
| Confabulations | 40.3% | 17.9% |
| LMArena Expert | 1303 | 1364 |
| SimpleQA Verified | — | 15.3% |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Not comparable
Gemma 3 27B: 32.0 (#94), o3-mini: —
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| LMArena Vision | 1165 | — |
| GeoBench | 52% | — |
Multilingual Gemma 3 27B leads
Gemma 3 27B: 47.0 (#150), o3-mini: 45.9 (#159)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| LMArena Non-English | 1334 | 1319 |
| LMArena Chinese | 1347 | 1379 |
| LMArena French | 1368 | 1333 |
| LMArena German | 1362 | 1303 |
| LMArena Japanese | 1287 | 1286 |
| LMArena Korean | 1308 | 1314 |
| LMArena Russian | 1349 | 1304 |
| LMArena Spanish | 1348 | 1321 |
Instruction Following o3-mini leads
Gemma 3 27B: 70.8 (#155), o3-mini: 75.3 (#67)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| LiveBench Instruction Following | 74.9% | 84.4% |
| LMArena Instruction Following | 1321 | 1338 |
Long Context o3-mini leads
Gemma 3 27B: 27.6 (#288), o3-mini: 33.8 (#251)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| Fiction.LiveBench | 33.3% | 50% |
| LMArena Longer Query | 1333 | 1343 |
Writing & Preference Gemma 3 27B leads
Gemma 3 27B: 52.7 (#161), o3-mini: 50.4 (#177)
| Benchmark | Gemma 3 27B | o3-mini |
|---|---|---|
| LMArena Text | 1358 | 1337 |
| LMArena Creative Writing | 1346 | 1286 |
| Short-Story Creative Writing | 79.9% | 61.7% |
| LMArena Multi-Turn | 1344 | 1320 |
| LiveBench Language | 34.6% | 50.7% |
| EQ-Bench Creative Writing | 1266 | — |
Frequently asked questions
Is Gemma 3 27B better than o3-mini?
o3-mini is the stronger model overall, scoring 36.7 to 30.5 on the Noometry Index. Gemma 3 27B costs 19× less per token, which makes it the better buy when o3-mini's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 27B or o3-mini?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is Gemma 3 27B or o3-mini better for coding?
o3-mini scores higher on coding benchmarks: 41.4 versus 22.1 in the Noometry coding category.
Which has the bigger context window?
o3-mini does, with 200K tokens against 131K.
How many benchmarks do Gemma 3 27B and o3-mini share?
37 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and o3-mini has 51.