Model comparison
Grok 4.7 vs MiniMax-M3
Grok 4.7 is the stronger model overall, scoring 53.1 to 43.8 on the Noometry Index. MiniMax-M3 costs 5.7× less per token, which makes it the better buy when Grok 4.7's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
MiniMaxSummary
- They share 30 benchmarks with published results for both. Grok 4.7 scores higher in 6 categories and MiniMax-M3 in 4 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.7 leads 49.1 to 30.1.
- The biggest single-benchmark swing is FrontierCode: 47.6% for Grok 4.7 and 14.7% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $2 / $6 for Grok 4.7.
- MiniMax-M3 accepts more context: 1M tokens versus 500K.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| Grok 4.7 | MiniMax-M3 | |
|---|---|---|
| Provider | xAI | MiniMax |
| Noometry Index | 53.1 | 43.8 |
| Released | 2026-09-21 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | 500K | 1M |
| Max output | 500K | 512K |
| Input $ / M tokens | $2 | $0.30 |
| Output $ / M tokens | $6 | $1.20 |
| Results tracked | 39 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.7 leads
Grok 4.7: 58.0 (#18), MiniMax-M3: 41.8 (#118)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| FrontierCode | 47.6% | 14.7% |
| LMArena WebDev | 1639 | 1482 |
| SciCode | 57.8% | 47.1% |
| LMArena Coding | 1427 | 1469 |
| CursorBench | 46.3% | — |
| FrontierSWE | 29.5% | — |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use Grok 4.7 leads
Grok 4.7: 36.7 (#37), MiniMax-M3: 22.6 (#130)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| APEX-Agents | 54.6% | 37.7% |
| Vending-Bench 2 | 10,537 | 2,158 |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| GDP.pdf | 22.8% | — |
Reasoning Grok 4.7 leads
Grok 4.7: 49.1 (#40), MiniMax-M3: 30.1 (#87)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| NYT Connections (extended) | 76.8% | 65.1% |
| CritPt | 18% | 3.7% |
| Chess Puzzles | 38% | 14% |
| LMArena Hard Prompts | 1413 | 1447 |
| Mystery Game Puzzles | 29% | 8% |
| DTBench | 96% | 78.9% |
| LMCA | 49.4% | 33.7% |
| Epoch Capabilities Index | 153.53 | 146.95 |
| SimpleBench | — | 45.8% |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
Math Grok 4.7 leads
Grok 4.7: 57.8 (#39), MiniMax-M3: 40.0 (#95)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.1% | 71.1% |
| ProofBench | 34% | 18% |
| LMArena Math | 1407 | 1429 |
| FrontierMath (Tiers 1-3) | 53% | — |
| FrontierMath Tier 4 | 17.1% | — |
Knowledge Grok 4.7 leads
Grok 4.7: 62.8 (#22), MiniMax-M3: 58.4 (#35)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 92.7% | 90.9% |
| LMArena Expert | 1422 | 1461 |
| SimpleQA Verified | 56% | — |
Multimodal MiniMax-M3 leads
Grok 4.7: 35.5 (#87), MiniMax-M3: 40.2 (#51)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1228 | 1253 |
| Blueprint-Bench 2 | 32.5% | — |
| Furniture Assembly | 20.8% | — |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
Grok 4.7: 50.8 (#116), MiniMax-M3: 53.0 (#75)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1389 | 1420 |
| LMArena Chinese | 1455 | 1463 |
| LMArena French | 1455 | 1447 |
| LMArena Russian | 1397 | 1428 |
| LMArena Spanish | 1400 | 1432 |
| LMArena German | — | 1426 |
| LMArena Japanese | — | 1381 |
| LMArena Korean | — | 1372 |
Instruction Following MiniMax-M3 leads
Grok 4.7: 74.1 (#105), MiniMax-M3: 75.5 (#62)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1433 |
Long Context MiniMax-M3 leads
Grok 4.7: 43.1 (#104), MiniMax-M3: 44.2 (#72)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1413 | 1445 |
Writing & Preference Grok 4.7 leads
Grok 4.7: 70.0 (#24), MiniMax-M3: 62.1 (#83)
| Benchmark | Grok 4.7 | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1399 | 1433 |
| LMArena Creative Writing | 1391 | 1404 |
| LMArena Multi-Turn | 1393 | 1442 |
| EQ-Bench Creative Writing | 2007 | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Grok 4.7 better than MiniMax-M3?
Grok 4.7 is the stronger model overall, scoring 53.1 to 43.8 on the Noometry Index. MiniMax-M3 costs 5.7× less per token, which makes it the better buy when Grok 4.7's lead doesn't matter for your workload.
Which is cheaper, Grok 4.7 or MiniMax-M3?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Grok 4.7 lists at $2 and $6.
Is Grok 4.7 or MiniMax-M3 better for coding?
Grok 4.7 scores higher on coding benchmarks: 58.0 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 500K.
How many benchmarks do Grok 4.7 and MiniMax-M3 share?
30 benchmarks have published results for both models. Grok 4.7 has 39 scored results on Noometry and MiniMax-M3 has 41.