Model comparison
Gemma 3 4B vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.8 to 28.1 on the Noometry Index. Gemma 3 4B costs 10× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
Google
MiniMaxSummary
- They share 18 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and MiniMax-M3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 11.8.
- The biggest single-benchmark swing is GPQA Diamond: 23.2% for Gemma 3 4B and 90.9% for MiniMax-M3.
- Gemma 3 4B is cheaper at $0.04 / $0.08 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 4B | MiniMax-M3 | |
|---|---|---|
| Provider | MiniMax | |
| Noometry Index | 28.1 | 43.8 |
| Released | 2025-03-12 | 2026-06-01 |
| Weights | Open | Open |
| Context window | 131K | 1M |
| Max output | 4K | 512K |
| Input $ / M tokens | $0.04 | $0.30 |
| Output $ / M tokens | $0.08 | $1.20 |
| Results tracked | 22 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
Gemma 3 4B: 35.9 (#215), MiniMax-M3: 41.8 (#118)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| LMArena Coding | 1230 | 1469 |
| FrontierCode | — | 14.7% |
| LMArena WebDev | — | 1482 |
| SciCode | — | 47.1% |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use MiniMax-M3 leads
Gemma 3 4B: 20.9 (#142), MiniMax-M3: 22.6 (#130)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| Berkeley Function Calling Leaderboard | 19.6% | — |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
Gemma 3 4B: 13.2 (#335), MiniMax-M3: 30.1 (#87)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| Chess Puzzles | 0% | 14% |
| LMArena Hard Prompts | 1253 | 1447 |
| DTBench | 50.9% | 78.9% |
| LMCA | 2.8% | 33.7% |
| Epoch Capabilities Index | 116.02 | 146.95 |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | 25.2% | — |
| NYT Connections (extended) | — | 65.1% |
| CritPt | — | 3.7% |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
Math MiniMax-M3 leads
Gemma 3 4B: 16.8 (#292), MiniMax-M3: 40.0 (#95)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.5% | 71.1% |
| LMArena Math | 1239 | 1429 |
| ProofBench | — | 18% |
Knowledge MiniMax-M3 leads
Gemma 3 4B: 11.8 (#299), MiniMax-M3: 58.4 (#35)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 23.2% | 90.9% |
| LMArena Expert | 1223 | 1461 |
| Vectara Hallucination Rate | 6.4% | — |
Multimodal Not comparable
Gemma 3 4B: —, MiniMax-M3: 40.2 (#51)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1253 |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
Gemma 3 4B: 42.5 (#194), MiniMax-M3: 53.0 (#75)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1273 | 1420 |
| LMArena German | 1281 | 1426 |
| LMArena Russian | 1294 | 1428 |
| LMArena Chinese | — | 1463 |
| LMArena French | — | 1447 |
| LMArena Japanese | — | 1381 |
| LMArena Korean | — | 1372 |
| LMArena Spanish | — | 1432 |
Instruction Following MiniMax-M3 leads
Gemma 3 4B: 65.2 (#225), MiniMax-M3: 75.5 (#62)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1239 | 1433 |
Long Context MiniMax-M3 leads
Gemma 3 4B: 38.7 (#194), MiniMax-M3: 44.2 (#72)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1273 | 1445 |
Writing & Preference MiniMax-M3 leads
Gemma 3 4B: 42.0 (#239), MiniMax-M3: 62.1 (#83)
| Benchmark | Gemma 3 4B | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1291 | 1433 |
| LMArena Creative Writing | 1271 | 1404 |
| LMArena Multi-Turn | 1255 | 1442 |
| EQ-Bench Creative Writing | 1068 | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Gemma 3 4B better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 28.1 on the Noometry Index. Gemma 3 4B costs 10× 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 4B or MiniMax-M3?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is Gemma 3 4B or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 35.9 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 4B and MiniMax-M3 share?
18 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and MiniMax-M3 has 41.