Model comparison
Gemma 3 27B vs o3
o3 is the stronger model overall, scoring 47.5 to 30.8 on the Noometry Index. Gemma 3 27B costs 35× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
GoogleSummary
- They share 34 benchmarks with published results for both. Gemma 3 27B scores higher in 0 categories and o3 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o3 leads 54.6 to 25.5.
- The biggest single-benchmark swing is Aider Polyglot: 4.9% for Gemma 3 27B and 81.3% for o3.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $2 / $8 for o3.
- o3 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 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 30.8 | 47.5 |
| Released | 2025-03-11 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 131K | 200K |
| Max output | 8K | 100K |
| Input $ / M tokens | $0.08 | $2 |
| Output $ / M tokens | $0.16 | $8 |
| Results tracked | 43 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
Gemma 3 27B: 22.5 (#334), o3: 46.8 (#64)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| Aider Polyglot | 4.9% | 81.3% |
| LMArena Coding | 1322 | 1408 |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| SciCode | 21.2% | — |
| GSO | — | 8.8% |
| WeirdML | — | 52.4% |
| LiveBench Coding | 39.9% | — |
| CadEval | — | 74% |
| ALE-Bench | — | 933.55 |
Agentic & Tool Use o3 leads
Gemma 3 27B: 25.1 (#110), o3: 34.5 (#44)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
Gemma 3 27B: 16.7 (#301), o3: 32.0 (#78)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| Kagi LLM Benchmark | 40.4% | 67.6% |
| CritPt | 0% | 1.4% |
| Chess Puzzles | 0% | 38% |
| LMArena Hard Prompts | 1340 | 1402 |
| DTBench | 52.5% | 84.8% |
| LMCA | 12.3% | 39.7% |
| Epoch Capabilities Index | 130.04 | 146.86 |
| ARC-AGI-2 | — | 6.5% |
| SimpleBench | — | 53.1% |
| ARC-AGI-1 | — | 60.8% |
| EnigmaEval | — | 13.1% |
| LiveBench Reasoning | 43.8% | — |
| Mystery Game Puzzles | — | 29% |
| LiveBench Data Analysis | 51.5% | — |
| ForecastBench | — | 62.5 |
| LiveBench | 50% | — |
Math o3 leads
Gemma 3 27B: 25.9 (#265), o3: 50.2 (#58)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 84.4% |
| LMArena Math | 1312 | 1426 |
| MATH Level 5 | 74% | 97.8% |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| Omni-MATH | — | 71.4% |
| LiveBench Math | 55.4% | — |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
Gemma 3 27B: 25.5 (#261), o3: 54.6 (#52)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| GPQA Diamond | 47.7% | 81.8% |
| Confabulations | 40.3% | 14.4% |
| LMArena Expert | 1304 | 1402 |
| Humanity's Last Exam | — | 20.3% |
| SimpleQA Verified | — | 49.4% |
| MMLU-Pro | — | 85.9% |
| Vectara Hallucination Rate | 7.4% | — |
| GPQA (HELM) | — | 75.3% |
Multimodal o3 leads
Gemma 3 27B: 32.6 (#100), o3: 41.4 (#36)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| LMArena Vision | 1164 | 1214 |
| GeoBench | 52% | 74% |
| VPCT | — | 52% |
Multilingual o3 leads
Gemma 3 27B: 46.9 (#155), o3: 51.7 (#105)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| LMArena Non-English | 1334 | 1401 |
| LMArena Chinese | 1346 | 1437 |
| LMArena French | 1368 | 1430 |
| LMArena German | 1362 | 1420 |
| LMArena Japanese | 1287 | 1403 |
| LMArena Korean | 1308 | 1370 |
| LMArena Russian | 1349 | 1406 |
| LMArena Spanish | 1349 | 1395 |
Instruction Following o3 leads
Gemma 3 27B: 70.6 (#160), o3: 72.8 (#127)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| LMArena Instruction Following | 1321 | 1368 |
| LiveBench Instruction Following | 74.9% | — |
| IFEval | — | 86.9% |
Long Context o3 leads
Gemma 3 27B: 27.6 (#293), o3: 53.3 (#6)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| Fiction.LiveBench | 33.3% | 88.9% |
| LMArena Longer Query | 1333 | 1372 |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
Gemma 3 27B: 52.5 (#168), o3: 63.5 (#64)
| Benchmark | Gemma 3 27B | o3 |
|---|---|---|
| LMArena Text | 1358 | 1410 |
| LMArena Creative Writing | 1346 | 1359 |
| Short-Story Creative Writing | 79.9% | 83.9% |
| EQ-Bench Creative Writing | 1266 | 1676 |
| LMArena Multi-Turn | 1345 | 1405 |
| WildBench | — | 86.1% |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than o3?
o3 is the stronger model overall, scoring 47.5 to 30.8 on the Noometry Index. Gemma 3 27B costs 35× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 27B or o3?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; o3 lists at $2 and $8.
Is Gemma 3 27B or o3 better for coding?
o3 scores higher on coding benchmarks: 46.8 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
o3 does, with 200K tokens against 131K.
How many benchmarks do Gemma 3 27B and o3 share?
34 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and o3 has 63.