Model comparison
Grok 3 vs MiniMax-M2.5
Grok 3 is the stronger model overall, scoring 39.6 to 38.1 on the Noometry Index.
Last verified . 23 shared benchmarks.
MiniMaxSummary
- They share 23 benchmarks with published results for both. Grok 3 scores higher in 7 categories and MiniMax-M2.5 in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 3 leads 36.2 to 25.8.
- The biggest single-benchmark swing is ARC-AGI-1: 5.5% for Grok 3 and 63.7% for MiniMax-M2.5.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | MiniMax-M2.5 | |
|---|---|---|
| Provider | xAI | MiniMax |
| Noometry Index | 39.6 | 38.1 |
| Released | 2025-04-09 | 2026-02-12 |
| Weights | Proprietary | Open |
| Context window | — | 205K |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 40 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Grok 3: 41.5 (#113), MiniMax-M2.5: 47.9 (#53)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| LMArena Coding | 1432 | 1380 |
| SWE-bench Verified (bash only) | — | 75.8% |
| Aider Polyglot | 53.3% | — |
| LMArena WebDev | — | 1386 |
| SWE-bench Multilingual | — | 68.3% |
| WeirdML | 37.2% | — |
| ALE-Bench | — | 618.17 |
Agentic & Tool Use Too close to call
Grok 3: 30.2 (#69), MiniMax-M2.5: 30.2 (#70)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| BALROG | 29.5% | — |
| Vending-Bench 2 | — | -23.16 |
Reasoning MiniMax-M2.5 leads
Grok 3: 13.8 (#322), MiniMax-M2.5: 17.6 (#281)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| ARC-AGI-2 | 0% | 4.9% |
| Kagi LLM Benchmark | 61.3% | 55.2% |
| ARC-AGI-1 | 5.5% | 63.7% |
| LMArena Hard Prompts | 1433 | 1372 |
| Epoch Capabilities Index | 138.34 | 146.68 |
| SimpleBench | 36.1% | — |
| NYT Connections (extended) | — | 16.8% |
Math Grok 3 leads
Grok 3: 36.2 (#122), MiniMax-M2.5: 25.8 (#242)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1390 | 1378 |
| OTIS Mock AIME 2024-2025 | 55.6% | — |
| ProofBench | — | 4% |
| Omni-MATH | 46.4% | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Grok 3 leads
Grok 3: 46.0 (#76), MiniMax-M2.5: 38.6 (#132)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | 5.8% | 9.1% |
| LMArena Expert | 1420 | 1378 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| GPQA (HELM) | 65% | — |
Multilingual Grok 3 leads
Grok 3: 52.5 (#81), MiniMax-M2.5: 47.3 (#147)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1411 | 1338 |
| LMArena Chinese | 1448 | 1392 |
| LMArena French | 1460 | 1363 |
| LMArena German | 1431 | 1361 |
| LMArena Japanese | 1389 | 1169 |
| LMArena Korean | 1374 | 1232 |
| LMArena Russian | 1416 | 1359 |
| LMArena Spanish | 1417 | 1354 |
Instruction Following Grok 3 leads
Grok 3: 75.2 (#69), MiniMax-M2.5: 71.7 (#143)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1409 | 1353 |
| IFEval | 88.4% | — |
Long Context Grok 3 leads
Grok 3: 38.7 (#187), MiniMax-M2.5: 37.4 (#211)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1439 | 1367 |
| Fiction.LiveBench | 58.3% | — |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference Grok 3 leads
Grok 3: 56.0 (#136), MiniMax-M2.5: 54.1 (#149)
| Benchmark | Grok 3 | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1426 | 1359 |
| LMArena Creative Writing | 1414 | 1331 |
| EQ-Bench Creative Writing | 1186 | 1361 |
| LMArena Multi-Turn | 1424 | 1364 |
| Short-Story Creative Writing | 76.4% | — |
| WildBench | 84.9% | — |
Frequently asked questions
Is Grok 3 better than MiniMax-M2.5?
Grok 3 is the stronger model overall, scoring 39.6 to 38.1 on the Noometry Index.
Is Grok 3 or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 41.5 in the Noometry coding category.
How many benchmarks do Grok 3 and MiniMax-M2.5 share?
23 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and MiniMax-M2.5 has 33.