Model comparison
Gemma 4 26B A4B IT vs o3-mini
Gemma 4 26B A4B IT is the stronger model overall, scoring 43.5 to 36.7 on the Noometry Index.
Last verified . 23 shared benchmarks.
GoogleSummary
- They share 23 benchmarks with published results for both. Gemma 4 26B A4B IT scores higher in 6 categories and o3-mini in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 26B A4B IT leads 47.6 to 28.1.
- The biggest single-benchmark swing is Chess Puzzles: 6% for Gemma 4 26B A4B IT and 17% for o3-mini.
- Gemma 4 26B A4B IT is cheaper at $0.0675 / $0.23 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- Gemma 4 26B A4B IT accepts more context: 262K tokens versus 200K.
- Gemma 4 26B A4B IT has downloadable open weights; the other is API-only.
Side by side
| Gemma 4 26B A4B IT | o3-mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 36.7 |
| Released | 2026-04-02 | 2024-12-20 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 33K | 100K |
| Input $ / M tokens | $0.0675 | $1.10 |
| Output $ / M tokens | $0.23 | $4.40 |
| Results tracked | 28 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-mini leads
Gemma 4 26B A4B IT: 39.0 (#164), o3-mini: 40.8 (#132)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| SciCode | 40% | 39.8% |
| WeirdML | 35.2% | 43.7% |
| LMArena Coding | 1447 | 1378 |
| Aider Polyglot | — | 60.4% |
| LMArena WebDev | 1359 | — |
| GSO | — | 1.3% |
| LiveBench Coding | — | 82.7% |
| CadEval | — | 54% |
| ALE-Bench | 927.17 | — |
Agentic & Tool Use Not comparable
Gemma 4 26B A4B IT: —, o3-mini: 29.6 (#84)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| Cybench | — | 22.5% |
Reasoning Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 21.8 (#213), o3-mini: 16.3 (#305)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| CritPt | 0% | 0.3% |
| Chess Puzzles | 6% | 17% |
| LMArena Hard Prompts | 1439 | 1366 |
| DTBench | 74.9% | 68.8% |
| LMCA | 29.7% | 19% |
| Epoch Capabilities Index | 141.85 | 140.34 |
| ARC-AGI-2 | — | 3% |
| SimpleBench | — | 22.8% |
| ARC-AGI-1 | — | 34.5% |
| LiveBench Reasoning | — | 89.6% |
| Mystery Game Puzzles | — | 7% |
| LiveBench Data Analysis | — | 70.6% |
| ForecastBench | — | 59.6 |
| LiveBench | — | 75.9% |
Math Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 47.6 (#67), o3-mini: 28.1 (#244)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 82.2% | 76.9% |
| LMArena Math | 1470 | 1396 |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| LiveBench Math | — | 77.3% |
| MATH Level 5 | — | 96.5% |
| FrontierMath (Feb 2025 set) | — | 12.4% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 45.7 (#85), o3-mini: 38.3 (#146)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| GPQA Diamond | 73.2% | 77% |
| LMArena Expert | 1447 | 1364 |
| SimpleQA Verified | — | 15.3% |
| Confabulations | — | 17.9% |
| Vectara Hallucination Rate | 5.2% | — |
Multimodal Not comparable
Gemma 4 26B A4B IT: 40.6 (#46), o3-mini: —
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| LMArena Vision | 1260 | — |
Multilingual Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 53.1 (#71), o3-mini: 45.7 (#164)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| LMArena Non-English | 1421 | 1319 |
| LMArena Chinese | 1495 | 1379 |
| LMArena French | 1460 | 1334 |
| LMArena Russian | 1434 | 1304 |
| LMArena Spanish | 1417 | 1321 |
| LMArena German | — | 1303 |
| LMArena Japanese | — | 1286 |
| LMArena Korean | — | 1314 |
Instruction Following Too close to call
Gemma 4 26B A4B IT: 74.8 (#82), o3-mini: 75.1 (#72)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1420 | 1337 |
| LiveBench Instruction Following | — | 84.4% |
Long Context Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 43.6 (#91), o3-mini: 33.8 (#256)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| LMArena Longer Query | 1428 | 1343 |
| Fiction.LiveBench | — | 50% |
Writing & Preference Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 58.6 (#115), o3-mini: 50.3 (#182)
| Benchmark | Gemma 4 26B A4B IT | o3-mini |
|---|---|---|
| LMArena Text | 1434 | 1337 |
| LMArena Creative Writing | 1402 | 1286 |
| LMArena Multi-Turn | 1441 | 1320 |
| Short-Story Creative Writing | — | 61.7% |
| EQ-Bench Creative Writing | 1305 | — |
| LiveBench Language | — | 50.7% |
Frequently asked questions
Is Gemma 4 26B A4B IT better than o3-mini?
Gemma 4 26B A4B IT is the stronger model overall, scoring 43.5 to 36.7 on the Noometry Index.
Which is cheaper, Gemma 4 26B A4B IT or o3-mini?
Gemma 4 26B A4B IT is cheaper. It lists at $0.0675 per million input tokens and $0.23 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is Gemma 4 26B A4B IT or o3-mini better for coding?
o3-mini scores higher on coding benchmarks: 40.8 versus 39.0 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 26B A4B IT does, with 262K tokens against 200K.
How many benchmarks do Gemma 4 26B A4B IT and o3-mini share?
23 benchmarks have published results for both models. Gemma 4 26B A4B IT has 28 scored results on Noometry and o3-mini has 51.