Model comparison
GPT-6 Sol vs MiniMax-M3
GPT-6 Sol is the stronger model overall, scoring 61.8 to 43.8 on the Noometry Index. MiniMax-M3 costs 7.6× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
MiniMaxSummary
- They share 33 benchmarks with published results for both. GPT-6 Sol scores higher in 7 categories and MiniMax-M3 in 3 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Sol leads 87.2 to 40.0.
- The biggest single-benchmark swing is ProofBench: 83% for GPT-6 Sol and 18% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 1M.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Sol | MiniMax-M3 | |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 61.8 | 43.8 |
| Released | 2026-09-22 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 1M |
| Max output | 128K | 512K |
| Input $ / M tokens | $2 | $0.30 |
| Output $ / M tokens | $10 | $1.20 |
| Results tracked | 45 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
GPT-6 Sol: 60.1 (#11), MiniMax-M3: 41.8 (#118)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| FrontierCode | 49.3% | 14.7% |
| LMArena WebDev | 1688 | 1482 |
| SciCode | 57.6% | 47.1% |
| LMArena Coding | 1447 | 1469 |
| ALE-Bench | 2,462 | 640.02 |
| DeepSWE | 68.8% | — |
Agentic & Tool Use GPT-6 Sol leads
GPT-6 Sol: 37.2 (#36), MiniMax-M3: 22.6 (#130)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| APEX-Agents | 54.3% | 37.7% |
| Vending-Bench 2 | 14,428 | 2,158 |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| GDP.pdf | 26.4% | — |
Reasoning GPT-6 Sol leads
GPT-6 Sol: 74.0 (#9), MiniMax-M3: 30.1 (#87)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| NYT Connections (extended) | 90.1% | 65.1% |
| CritPt | 30.9% | 3.7% |
| LMArena Hard Prompts | 1418 | 1447 |
| Mystery Game Puzzles | 56% | 8% |
| DTBench | 97.3% | 78.9% |
| LMCA | 59.1% | 33.7% |
| Epoch Capabilities Index | 162.72 | 146.95 |
| ARC-AGI-2 | 89.6% | — |
| SimpleBench | — | 45.8% |
| ARC-AGI-1 | 95.5% | — |
| Chess Puzzles | — | 14% |
| EBR-Bench | 53.3% | — |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
Math GPT-6 Sol leads
GPT-6 Sol: 87.2 (#7), MiniMax-M3: 40.0 (#95)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 71.1% |
| ProofBench | 83% | 18% |
| LMArena Math | 1402 | 1429 |
| FrontierMath (Tiers 1-3) | 89.8% | — |
| FrontierMath Tier 4 | 90% | — |
Knowledge GPT-6 Sol leads
GPT-6 Sol: 64.8 (#15), MiniMax-M3: 58.4 (#35)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 94.3% | 90.9% |
| LMArena Expert | 1439 | 1461 |
| SimpleQA Verified | 60.7% | — |
| Vectara Hallucination Rate | 6.5% | — |
Multimodal GPT-6 Sol leads
GPT-6 Sol: 47.6 (#10), MiniMax-M3: 40.2 (#51)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1245 | 1253 |
| Blueprint-Bench 2 | 36.9% | — |
| Furniture Assembly | 58.3% | — |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
GPT-6 Sol: 50.5 (#118), MiniMax-M3: 53.0 (#75)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1385 | 1420 |
| LMArena Chinese | 1405 | 1463 |
| LMArena French | 1410 | 1447 |
| LMArena German | 1390 | 1426 |
| LMArena Japanese | 1385 | 1381 |
| LMArena Korean | 1341 | 1372 |
| LMArena Russian | 1401 | 1428 |
| LMArena Spanish | 1384 | 1432 |
Instruction Following MiniMax-M3 leads
GPT-6 Sol: 74.5 (#94), MiniMax-M3: 75.5 (#62)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1412 | 1433 |
Long Context MiniMax-M3 leads
GPT-6 Sol: 43.1 (#108), MiniMax-M3: 44.2 (#72)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1411 | 1445 |
Writing & Preference GPT-6 Sol leads
GPT-6 Sol: 71.9 (#18), MiniMax-M3: 62.1 (#83)
| Benchmark | GPT-6 Sol | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1395 | 1433 |
| LMArena Creative Writing | 1378 | 1404 |
| LMArena Multi-Turn | 1412 | 1442 |
| EQ-Bench Creative Writing | 2125 | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is GPT-6 Sol better than MiniMax-M3?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 43.8 on the Noometry Index. MiniMax-M3 costs 7.6× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.
Which is cheaper, GPT-6 Sol or MiniMax-M3?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-6 Sol lists at $2 and $10.
Is GPT-6 Sol or MiniMax-M3 better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 1M.
How many benchmarks do GPT-6 Sol and MiniMax-M3 share?
33 benchmarks have published results for both models. GPT-6 Sol has 45 scored results on Noometry and MiniMax-M3 has 41.