Model comparison
Gemma 3 12B vs MiniMax-M2.7
MiniMax-M2.7 is the stronger model overall, scoring 37.4 to 31.8 on the Noometry Index. Gemma 3 12B costs 7.0× less per token, which makes it the better buy when MiniMax-M2.7's lead doesn't matter for your workload.
Last verified . 16 shared benchmarks.
Google
MiniMaxSummary
- They share 16 benchmarks with published results for both. Gemma 3 12B scores higher in 1 category and MiniMax-M2.7 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.7 leads 59.1 to 47.7.
- The biggest single-benchmark swing is SciCode: 17.4% for Gemma 3 12B and 47% for MiniMax-M2.7.
- Gemma 3 12B is cheaper at $0.05 / $0.15 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.7.
- MiniMax-M2.7 accepts more context: 205K tokens versus 131K.
Side by side
| Gemma 3 12B | MiniMax-M2.7 | |
|---|---|---|
| Provider | MiniMax | |
| Noometry Index | 31.8 | 37.4 |
| Released | 2025-03-12 | 2026-03-18 |
| Weights | Open | Open |
| Context window | 131K | 205K |
| Max output | 8K | 131K |
| Input $ / M tokens | $0.05 | $0.30 |
| Output $ / M tokens | $0.15 | $1.20 |
| Results tracked | 24 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.7 leads
Gemma 3 12B: 31.2 (#276), MiniMax-M2.7: 41.4 (#115)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| SciCode | 17.4% | 47% |
| LMArena Coding | 1281 | 1454 |
| LMArena WebDev | — | 1397 |
| WeirdML | — | 37% |
| ALE-Bench | — | 599.25 |
Agentic & Tool Use Too close to call
Gemma 3 12B: 25.3 (#101), MiniMax-M2.7: 24.9 (#103)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| Terminal-Bench | — | 45.1% |
| Berkeley Function Calling Leaderboard | 30.4% | — |
| ExploitBench | — | 13.3% |
| GBAEval | — | 0% |
Reasoning MiniMax-M2.7 leads
Gemma 3 12B: 15.2 (#307), MiniMax-M2.7: 19.0 (#248)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| CritPt | 0% | 0.6% |
| LMArena Hard Prompts | 1309 | 1422 |
| Epoch Capabilities Index | 123.49 | 145.84 |
| NYT Connections (extended) | — | 24.7% |
| Chess Puzzles | 0% | — |
| Thematic Generalization | — | 39.3% |
| DTBench | 48.8% | — |
| LMCA | 4.5% | — |
Math MiniMax-M2.7 leads
Gemma 3 12B: 21.0 (#273), MiniMax-M2.7: 24.9 (#254)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| LMArena Math | 1307 | 1420 |
| OTIS Mock AIME 2024-2025 | 16.7% | — |
| ProofBench | — | 3% |
Knowledge MiniMax-M2.7 leads
Gemma 3 12B: 26.5 (#250), MiniMax-M2.7: 37.1 (#151)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| Vectara Hallucination Rate | 4.4% | 12.9% |
| LMArena Expert | 1248 | 1443 |
| GPQA Diamond | 39.5% | — |
Multimodal Not comparable
Gemma 3 12B: —, MiniMax-M2.7: —
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| MindCube | 46.7% | — |
Multilingual MiniMax-M2.7 leads
Gemma 3 12B: 45.9 (#160), MiniMax-M2.7: 50.5 (#120)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| LMArena Non-English | 1318 | 1382 |
| LMArena German | 1370 | 1398 |
| LMArena Russian | 1335 | 1383 |
| LMArena Chinese | — | 1439 |
| LMArena French | — | 1422 |
| LMArena Japanese | — | 1261 |
| LMArena Korean | — | 1313 |
| LMArena Spanish | — | 1403 |
Instruction Following MiniMax-M2.7 leads
Gemma 3 12B: 68.8 (#179), MiniMax-M2.7: 74.2 (#101)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| LMArena Instruction Following | 1299 | 1404 |
Long Context MiniMax-M2.7 leads
Gemma 3 12B: 40.0 (#157), MiniMax-M2.7: 43.3 (#94)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| LMArena Longer Query | 1317 | 1419 |
Writing & Preference MiniMax-M2.7 leads
Gemma 3 12B: 47.7 (#205), MiniMax-M2.7: 59.1 (#107)
| Benchmark | Gemma 3 12B | MiniMax-M2.7 |
|---|---|---|
| LMArena Text | 1334 | 1404 |
| LMArena Creative Writing | 1331 | 1354 |
| LMArena Multi-Turn | 1334 | 1413 |
| EQ-Bench Creative Writing | 1126 | — |
Frequently asked questions
Is Gemma 3 12B better than MiniMax-M2.7?
MiniMax-M2.7 is the stronger model overall, scoring 37.4 to 31.8 on the Noometry Index. Gemma 3 12B costs 7.0× less per token, which makes it the better buy when MiniMax-M2.7's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 12B or MiniMax-M2.7?
Gemma 3 12B is cheaper. It lists at $0.05 per million input tokens and $0.15 per million output tokens; MiniMax-M2.7 lists at $0.30 and $1.20.
Is Gemma 3 12B or MiniMax-M2.7 better for coding?
MiniMax-M2.7 scores higher on coding benchmarks: 41.4 versus 31.2 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2.7 does, with 205K tokens against 131K.
How many benchmarks do Gemma 3 12B and MiniMax-M2.7 share?
16 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and MiniMax-M2.7 has 30.