Model comparison
Gemma 3 12B vs GPT-6 Luna
GPT-6 Luna is the stronger model overall, scoring 53.3 to 32.1 on the Noometry Index. Gemma 3 12B costs 2.7× less per token, which makes it the better buy when GPT-6 Luna's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
GoogleSummary
- They share 20 benchmarks with published results for both. Gemma 3 12B scores higher in 0 categories and GPT-6 Luna in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 22.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 16.7% for Gemma 3 12B and 98.9% for GPT-6 Luna.
- Gemma 3 12B is cheaper at $0.05 / $0.15 per million input/output tokens, against $0.10 / $0.50 for GPT-6 Luna.
- GPT-6 Luna accepts more context: 1.05M tokens versus 131K.
- Gemma 3 12B has downloadable open weights; the other is API-only.
Side by side
| Gemma 3 12B | GPT-6 Luna | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 32.1 | 53.3 |
| Released | 2025-03-12 | 2026-09-22 |
| Weights | Open | Proprietary |
| Context window | 131K | 1.05M |
| Max output | 8K | 128K |
| Input $ / M tokens | $0.05 | $0.10 |
| Output $ / M tokens | $0.15 | $0.50 |
| Results tracked | 24 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
Gemma 3 12B: 31.7 (#280), GPT-6 Luna: 55.5 (#25)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| SciCode | 17.4% | 54.6% |
| LMArena Coding | 1281 | 1439 |
| DeepSWE | — | 66.6% |
| FrontierCode | — | 42.4% |
| LMArena WebDev | — | 1581 |
| ALE-Bench | — | 1,577 |
Agentic & Tool Use GPT-6 Luna leads
Gemma 3 12B: 25.5 (#108), GPT-6 Luna: 33.3 (#54)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| APEX-Agents | — | 44.3% |
| Berkeley Function Calling Leaderboard | 30.4% | — |
| GDP.pdf | — | 23% |
Reasoning GPT-6 Luna leads
Gemma 3 12B: 15.7 (#313), GPT-6 Luna: 48.2 (#41)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| CritPt | 0% | 19.4% |
| Chess Puzzles | 0% | 31% |
| LMArena Hard Prompts | 1309 | 1411 |
| DTBench | 48.8% | 90.1% |
| LMCA | 4.5% | 44.5% |
| Epoch Capabilities Index | 123.5 | 156.28 |
| ARC-AGI-2 | — | 59.3% |
| NYT Connections (extended) | — | 68.7% |
| ARC-AGI-1 | — | 86.7% |
| Mystery Game Puzzles | — | 7% |
Math GPT-6 Luna leads
Gemma 3 12B: 22.3 (#279), GPT-6 Luna: 76.1 (#15)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 16.7% | 98.9% |
| LMArena Math | 1307 | 1416 |
| FrontierMath (Tiers 1-3) | — | 78.9% |
| FrontierMath Tier 4 | — | 56.1% |
| ProofBench | — | 64% |
Knowledge GPT-6 Luna leads
Gemma 3 12B: 26.5 (#257), GPT-6 Luna: 57.0 (#41)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| GPQA Diamond | 39.5% | 90.5% |
| LMArena Expert | 1248 | 1444 |
| SimpleQA Verified | — | 41.4% |
| Vectara Hallucination Rate | 4.4% | — |
Multimodal Not comparable
Gemma 3 12B: —, GPT-6 Luna: 42.4 (#30)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| LMArena Vision | — | 1217 |
| Blueprint-Bench 2 | — | 31.2% |
| Furniture Assembly | — | 44.2% |
| MindCube | 46.7% | — |
Multilingual GPT-6 Luna leads
Gemma 3 12B: 45.7 (#165), GPT-6 Luna: 50.5 (#117)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | 1318 | 1386 |
| LMArena German | 1370 | 1369 |
| LMArena Russian | 1335 | 1394 |
| LMArena Chinese | — | 1433 |
| LMArena French | — | 1420 |
| LMArena Japanese | — | 1369 |
| LMArena Korean | — | 1360 |
| LMArena Spanish | — | 1393 |
Instruction Following GPT-6 Luna leads
Gemma 3 12B: 68.6 (#186), GPT-6 Luna: 74.3 (#99)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | 1299 | 1409 |
Long Context GPT-6 Luna leads
Gemma 3 12B: 40.0 (#162), GPT-6 Luna: 43.0 (#111)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | 1317 | 1409 |
Writing & Preference GPT-6 Luna leads
Gemma 3 12B: 47.5 (#209), GPT-6 Luna: 58.3 (#119)
| Benchmark | Gemma 3 12B | GPT-6 Luna |
|---|---|---|
| LMArena Text | 1334 | 1391 |
| LMArena Creative Writing | 1331 | 1363 |
| LMArena Multi-Turn | 1334 | 1396 |
| EQ-Bench Creative Writing | 1126 | — |
Frequently asked questions
Is Gemma 3 12B better than GPT-6 Luna?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 32.1 on the Noometry Index. Gemma 3 12B costs 2.7× less per token, which makes it the better buy when GPT-6 Luna's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 12B or GPT-6 Luna?
Gemma 3 12B is cheaper. It lists at $0.05 per million input tokens and $0.15 per million output tokens; GPT-6 Luna lists at $0.10 and $0.50.
Is Gemma 3 12B or GPT-6 Luna better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 31.7 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 131K.
How many benchmarks do Gemma 3 12B and GPT-6 Luna share?
20 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and GPT-6 Luna has 42.