Model comparison
Grok 4.20 (Non-Reasoning) vs MiniMax-M3
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 43.5 on the Noometry Index. MiniMax-M3 costs 3.0× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.
Last verified . 31 shared benchmarks.
MiniMaxSummary
- They share 31 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 7 categories and MiniMax-M3 in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.2 to 29.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 92.2% for Grok 4.20 (Non-Reasoning) and 71.1% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 (Non-Reasoning).
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| Grok 4.20 (Non-Reasoning) | MiniMax-M3 | |
|---|---|---|
| Provider | xAI | MiniMax |
| Noometry Index | 48.5 | 43.5 |
| Released | 2026-02-17 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | 1M | 1M |
| Max output | 30K | 512K |
| Input $ / M tokens | $1.25 | $0.30 |
| Output $ / M tokens | $2.50 | $1.20 |
| Results tracked | 46 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Grok 4.20 (Non-Reasoning): 41.8 (#104), MiniMax-M3: 41.5 (#111)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| LMArena WebDev | 1374 | 1482 |
| LMArena Coding | 1458 | 1470 |
| ALE-Bench | 1,150 | 640.02 |
| FrontierCode | — | 14.7% |
| SciCode | — | 47.1% |
| WeirdML | 52.3% | — |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 34.3 (#43), MiniMax-M3: 22.5 (#121)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| Vending-Bench 2 | 4,663 | 2,158 |
| Terminal-Bench | 57.3% | — |
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| τ²-bench Banking | 18% | — |
| GBAEval | — | 0.9% |
| LMArena Search | 1189 | — |
Reasoning Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 52.2 (#30), MiniMax-M3: 29.3 (#83)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| NYT Connections (extended) | 85.4% | 65.1% |
| Chess Puzzles | 24% | 14% |
| LMArena Hard Prompts | 1451 | 1447 |
| DTBench | 90.1% | 78.9% |
| LMCA | 38.7% | 33.7% |
| Epoch Capabilities Index | 152.01 | 147 |
| ForecastBench | 61.4 | 61.4 |
| ARC-AGI-2 | 65.1% | — |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | 75% | — |
| ARC-AGI-1 | 89.5% | — |
| CritPt | — | 3.7% |
| Thematic Generalization | 63.8% | — |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
Math Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 48.0 (#54), MiniMax-M3: 39.1 (#83)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 71.1% |
| ProofBench | 14% | 18% |
| LMArena Math | 1455 | 1432 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 17.1% | — |
Knowledge MiniMax-M3 leads
Grok 4.20 (Non-Reasoning): 52.8 (#54), MiniMax-M3: 58.4 (#32)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 89.3% | 90.9% |
| LMArena Expert | 1440 | 1462 |
| SimpleQA Verified | 30.2% | — |
Multimodal MiniMax-M3 leads
Grok 4.20 (Non-Reasoning): 33.2 (#88), MiniMax-M3: 39.5 (#45)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1263 | 1254 |
| LMArena Document | 1416 | 1435 |
| Blueprint-Bench 2 | 0% | — |
Multilingual Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 54.7 (#38), MiniMax-M3: 53.2 (#70)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1440 | 1420 |
| LMArena Chinese | 1481 | 1465 |
| LMArena French | 1477 | 1445 |
| LMArena German | 1463 | 1424 |
| LMArena Japanese | 1450 | 1381 |
| LMArena Korean | 1417 | 1372 |
| LMArena Russian | 1457 | 1430 |
| LMArena Spanish | 1442 | 1431 |
Instruction Following Too close to call
Grok 4.20 (Non-Reasoning): 75.0 (#79), MiniMax-M3: 75.7 (#56)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1420 | 1433 |
Long Context Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 45.4 (#33), MiniMax-M3: 44.2 (#68)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1436 | 1445 |
| CL-bench | 22.2% | — |
| CL-bench Life | 11.9% | — |
Writing & Preference Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 65.9 (#42), MiniMax-M3: 62.3 (#79)
| Benchmark | Grok 4.20 (Non-Reasoning) | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1451 | 1433 |
| LMArena Creative Writing | 1439 | 1405 |
| LMArena Multi-Turn | 1456 | 1441 |
| EQ-Bench Creative Writing | 1574 | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Grok 4.20 (Non-Reasoning) better than MiniMax-M3?
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 43.5 on the Noometry Index. MiniMax-M3 costs 3.0× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.
Which is cheaper, Grok 4.20 (Non-Reasoning) or MiniMax-M3?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.
Is Grok 4.20 (Non-Reasoning) or MiniMax-M3 better for coding?
They score almost the same on coding (41.8 vs 41.5); test both on your own repository before choosing.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Grok 4.20 (Non-Reasoning) and MiniMax-M3 share?
31 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and MiniMax-M3 has 41.