Model comparison
Gemma 3 27B vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.5 to 30.5 on the Noometry Index. Gemma 3 27B costs 5.2× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Google
MiniMaxSummary
- They share 26 benchmarks with published results for both. Gemma 3 27B scores higher in 1 category and MiniMax-M3 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 25.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 22.5% for Gemma 3 27B and 71.1% for MiniMax-M3.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 accepts more context: 1M tokens versus 131K.
Side by side
| Gemma 3 27B | MiniMax-M3 | |
|---|---|---|
| Provider | MiniMax | |
| Noometry Index | 30.5 | 43.5 |
| Released | 2025-03-11 | 2026-06-01 |
| Weights | Open | Open |
| Context window | 131K | 1M |
| Max output | 8K | 512K |
| Input $ / M tokens | $0.08 | $0.30 |
| Output $ / M tokens | $0.16 | $1.20 |
| Results tracked | 43 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
Gemma 3 27B: 22.1 (#329), MiniMax-M3: 41.5 (#111)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| SciCode | 21.2% | 47.1% |
| LMArena Coding | 1322 | 1470 |
| FrontierCode | — | 14.7% |
| Aider Polyglot | 4.9% | — |
| LMArena WebDev | — | 1482 |
| LiveBench Coding | 39.9% | — |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use Gemma 3 27B leads
Gemma 3 27B: 24.9 (#105), MiniMax-M3: 22.5 (#121)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| Berkeley Function Calling Leaderboard | 29.5% | — |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
Gemma 3 27B: 16.3 (#297), MiniMax-M3: 29.3 (#83)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| CritPt | 0% | 3.7% |
| Chess Puzzles | 0% | 14% |
| LMArena Hard Prompts | 1340 | 1447 |
| DTBench | 52.5% | 78.9% |
| LMCA | 12.3% | 33.7% |
| Epoch Capabilities Index | 130.04 | 147 |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | 40.4% | — |
| NYT Connections (extended) | — | 65.1% |
| LiveBench Reasoning | 43.8% | — |
| Mystery Game Puzzles | — | 8% |
| LiveBench Data Analysis | 51.5% | — |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
| LiveBench | 50% | — |
Math MiniMax-M3 leads
Gemma 3 27B: 24.4 (#261), MiniMax-M3: 39.1 (#83)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 71.1% |
| LMArena Math | 1312 | 1432 |
| ProofBench | — | 18% |
| LiveBench Math | 55.4% | — |
| MATH Level 5 | 74% | — |
Knowledge MiniMax-M3 leads
Gemma 3 27B: 25.4 (#256), MiniMax-M3: 58.4 (#32)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 47.7% | 90.9% |
| LMArena Expert | 1303 | 1462 |
| Confabulations | 40.3% | — |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal MiniMax-M3 leads
Gemma 3 27B: 32.0 (#94), MiniMax-M3: 39.5 (#45)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1165 | 1254 |
| GeoBench | 52% | — |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
Gemma 3 27B: 47.0 (#150), MiniMax-M3: 53.2 (#70)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1334 | 1420 |
| LMArena Chinese | 1347 | 1465 |
| LMArena French | 1368 | 1445 |
| LMArena German | 1362 | 1424 |
| LMArena Japanese | 1287 | 1381 |
| LMArena Korean | 1308 | 1372 |
| LMArena Russian | 1349 | 1430 |
| LMArena Spanish | 1348 | 1431 |
Instruction Following MiniMax-M3 leads
Gemma 3 27B: 70.8 (#155), MiniMax-M3: 75.7 (#56)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1321 | 1433 |
| LiveBench Instruction Following | 74.9% | — |
Long Context MiniMax-M3 leads
Gemma 3 27B: 27.6 (#288), MiniMax-M3: 44.2 (#68)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1333 | 1445 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference MiniMax-M3 leads
Gemma 3 27B: 52.7 (#161), MiniMax-M3: 62.3 (#79)
| Benchmark | Gemma 3 27B | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1358 | 1433 |
| LMArena Creative Writing | 1346 | 1405 |
| LMArena Multi-Turn | 1344 | 1441 |
| Short-Story Creative Writing | 79.9% | — |
| EQ-Bench Creative Writing | 1266 | — |
| EQ-Bench 4 | — | 1150 |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.5 to 30.5 on the Noometry Index. Gemma 3 27B costs 5.2× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 27B or MiniMax-M3?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is Gemma 3 27B or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 22.1 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 131K.
How many benchmarks do Gemma 3 27B and MiniMax-M3 share?
26 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and MiniMax-M3 has 41.