Model comparison
Grok 4.20 Multi-Agent vs MiniMax-M3
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 43.8 on the Noometry Index. MiniMax-M3 costs 3.0× less per token, which makes it the better buy when Grok 4.20 Multi-Agent's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
MiniMaxSummary
- They share 19 benchmarks with published results for both. Grok 4.20 Multi-Agent scores higher in 5 categories and MiniMax-M3 in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 40.4.
- The biggest single-benchmark swing is NYT Connections (extended): 89.6% for Grok 4.20 Multi-Agent and 65.1% 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 Multi-Agent.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| Grok 4.20 Multi-Agent | MiniMax-M3 | |
|---|---|---|
| Provider | xAI | MiniMax |
| Noometry Index | 46.2 | 43.8 |
| Released | 2026-03-09 | 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 | 20 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 43.0 (#92), MiniMax-M3: 41.8 (#118)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| LMArena Coding | 1457 | 1469 |
| FrontierCode | — | 14.7% |
| LMArena WebDev | — | 1482 |
| SciCode | — | 47.1% |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use Not comparable
Grok 4.20 Multi-Agent: —, MiniMax-M3: 22.6 (#130)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| LMArena Search | 1204 | — |
| Vending-Bench 2 | — | 2,158 |
Reasoning Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 43.9 (#48), MiniMax-M3: 30.1 (#87)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| NYT Connections (extended) | 89.6% | 65.1% |
| LMArena Hard Prompts | 1448 | 1447 |
| SimpleBench | — | 45.8% |
| CritPt | — | 3.7% |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| DTBench | — | 78.9% |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
| Epoch Capabilities Index | — | 146.95 |
| ForecastBench | — | 61.4 |
Math Too close to call
Grok 4.20 Multi-Agent: 39.4 (#104), MiniMax-M3: 40.0 (#95)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| LMArena Math | 1442 | 1429 |
| OTIS Mock AIME 2024-2025 | — | 71.1% |
| ProofBench | — | 18% |
Knowledge MiniMax-M3 leads
Grok 4.20 Multi-Agent: 40.4 (#119), MiniMax-M3: 58.4 (#35)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| LMArena Expert | 1445 | 1461 |
| GPQA Diamond | — | 90.9% |
Multimodal Too close to call
Grok 4.20 Multi-Agent: 40.5 (#48), MiniMax-M3: 40.2 (#51)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1259 | 1253 |
| LMArena Document | — | 1435 |
Multilingual Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 54.4 (#43), MiniMax-M3: 53.0 (#75)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1440 | 1420 |
| LMArena Chinese | 1475 | 1463 |
| LMArena French | 1466 | 1447 |
| LMArena German | 1456 | 1426 |
| LMArena Japanese | 1405 | 1381 |
| LMArena Korean | 1416 | 1372 |
| LMArena Russian | 1457 | 1428 |
| LMArena Spanish | 1447 | 1432 |
Instruction Following Too close to call
Grok 4.20 Multi-Agent: 74.8 (#84), MiniMax-M3: 75.5 (#62)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1420 | 1433 |
Long Context Too close to call
Grok 4.20 Multi-Agent: 43.7 (#88), MiniMax-M3: 44.2 (#72)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1431 | 1445 |
Writing & Preference Grok 4.20 Multi-Agent leads
Grok 4.20 Multi-Agent: 64.0 (#59), MiniMax-M3: 62.1 (#83)
| Benchmark | Grok 4.20 Multi-Agent | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1450 | 1433 |
| LMArena Creative Writing | 1436 | 1404 |
| LMArena Multi-Turn | 1452 | 1442 |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Grok 4.20 Multi-Agent better than MiniMax-M3?
Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 43.8 on the Noometry Index. MiniMax-M3 costs 3.0× less per token, which makes it the better buy when Grok 4.20 Multi-Agent's lead doesn't matter for your workload.
Which is cheaper, Grok 4.20 Multi-Agent 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 Multi-Agent lists at $1.25 and $2.50.
Is Grok 4.20 Multi-Agent or MiniMax-M3 better for coding?
Grok 4.20 Multi-Agent scores higher on coding benchmarks: 43.0 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Grok 4.20 Multi-Agent and MiniMax-M3 share?
19 benchmarks have published results for both models. Grok 4.20 Multi-Agent has 20 scored results on Noometry and MiniMax-M3 has 41.