Model comparison
Grok 3 vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.8 to 39.9 on the Noometry Index.
Last verified . 21 shared benchmarks.
MiniMaxSummary
- They share 21 benchmarks with published results for both. Grok 3 scores higher in 2 categories and MiniMax-M3 in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where MiniMax-M3 leads 30.1 to 13.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 55.6% for Grok 3 and 71.1% for MiniMax-M3.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | MiniMax-M3 | |
|---|---|---|
| Provider | xAI | MiniMax |
| Noometry Index | 39.9 | 43.8 |
| Released | 2025-04-09 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 512K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 40 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Grok 3: 41.9 (#115), MiniMax-M3: 41.8 (#118)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| LMArena Coding | 1432 | 1469 |
| FrontierCode | — | 14.7% |
| Aider Polyglot | 53.3% | — |
| LMArena WebDev | — | 1482 |
| SciCode | — | 47.1% |
| WeirdML | 37.2% | — |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use Grok 3 leads
Grok 3: 30.5 (#76), MiniMax-M3: 22.6 (#130)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| BALROG | 29.5% | — |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
Grok 3: 13.7 (#333), MiniMax-M3: 30.1 (#87)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| SimpleBench | 36.1% | 45.8% |
| LMArena Hard Prompts | 1434 | 1447 |
| Epoch Capabilities Index | 138.33 | 146.95 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 61.3% | — |
| NYT Connections (extended) | — | 65.1% |
| ARC-AGI-1 | 5.5% | — |
| CritPt | — | 3.7% |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| DTBench | — | 78.9% |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
Math MiniMax-M3 leads
Grok 3: 38.0 (#145), MiniMax-M3: 40.0 (#95)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | 71.1% |
| LMArena Math | 1391 | 1429 |
| ProofBench | — | 18% |
| Omni-MATH | 46.4% | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge MiniMax-M3 leads
Grok 3: 46.2 (#82), MiniMax-M3: 58.4 (#35)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 75.8% | 90.9% |
| LMArena Expert | 1421 | 1461 |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |
Multimodal Not comparable
Grok 3: —, MiniMax-M3: 40.2 (#51)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1253 |
| LMArena Document | — | 1435 |
Multilingual Too close to call
Grok 3: 52.3 (#87), MiniMax-M3: 53.0 (#75)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1410 | 1420 |
| LMArena Chinese | 1448 | 1463 |
| LMArena French | 1460 | 1447 |
| LMArena German | 1431 | 1426 |
| LMArena Japanese | 1387 | 1381 |
| LMArena Korean | 1373 | 1372 |
| LMArena Russian | 1416 | 1428 |
| LMArena Spanish | 1417 | 1432 |
Instruction Following Too close to call
Grok 3: 75.0 (#73), MiniMax-M3: 75.5 (#62)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1409 | 1433 |
| IFEval | 88.4% | — |
Long Context MiniMax-M3 leads
Grok 3: 38.7 (#192), MiniMax-M3: 44.2 (#72)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1439 | 1445 |
| Fiction.LiveBench | 58.3% | — |
Writing & Preference MiniMax-M3 leads
Grok 3: 55.8 (#141), MiniMax-M3: 62.1 (#83)
| Benchmark | Grok 3 | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1426 | 1433 |
| LMArena Creative Writing | 1414 | 1404 |
| LMArena Multi-Turn | 1425 | 1442 |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Grok 3 better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 39.9 on the Noometry Index.
Is Grok 3 or MiniMax-M3 better for coding?
They score almost the same on coding (41.9 vs 41.8); test both on your own repository before choosing.
How many benchmarks do Grok 3 and MiniMax-M3 share?
21 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and MiniMax-M3 has 41.