Model comparison
GPT-4.1 nano vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.8 to 27.9 on the Noometry Index. GPT-4.1 nano costs 3.0× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
MiniMaxSummary
- They share 22 benchmarks with published results for both. GPT-4.1 nano scores higher in 1 category and MiniMax-M3 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 21.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 28.9% for GPT-4.1 nano and 71.1% for MiniMax-M3.
- GPT-4.1 nano is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- GPT-4.1 nano accepts more context: 1.05M tokens versus 1M.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| GPT-4.1 nano | MiniMax-M3 | |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 27.9 | 43.8 |
| Released | 2025-04-14 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 1M |
| Max output | 33K | 512K |
| Input $ / M tokens | $0.10 | $0.30 |
| Output $ / M tokens | $0.40 | $1.20 |
| Results tracked | 38 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
GPT-4.1 nano: 24.1 (#330), MiniMax-M3: 41.8 (#118)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| SciCode | 25.9% | 47.1% |
| LMArena Coding | 1306 | 1469 |
| FrontierCode | — | 14.7% |
| Aider Polyglot | 8.9% | — |
| LMArena WebDev | — | 1482 |
| WeirdML | 19% | — |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use GPT-4.1 nano leads
GPT-4.1 nano: 26.5 (#104), MiniMax-M3: 22.6 (#130)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| Berkeley Function Calling Leaderboard | 33% | — |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
GPT-4.1 nano: 8.5 (#349), MiniMax-M3: 30.1 (#87)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| CritPt | 0% | 3.7% |
| LMArena Hard Prompts | 1286 | 1447 |
| DTBench | 52.5% | 78.9% |
| LMCA | 5.5% | 33.7% |
| Epoch Capabilities Index | 129.62 | 146.95 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | 33.3% | — |
| NYT Connections (extended) | — | 65.1% |
| ARC-AGI-1 | 0% | — |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
Math MiniMax-M3 leads
GPT-4.1 nano: 26.9 (#252), MiniMax-M3: 40.0 (#95)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 28.9% | 71.1% |
| LMArena Math | 1274 | 1429 |
| ProofBench | — | 18% |
| Omni-MATH | 36.7% | — |
| MATH Level 5 | 70% | — |
| FrontierMath (Feb 2025 set) | 1% | — |
Knowledge MiniMax-M3 leads
GPT-4.1 nano: 21.8 (#273), MiniMax-M3: 58.4 (#35)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 48.9% | 90.9% |
| LMArena Expert | 1272 | 1461 |
| SimpleQA Verified | 6% | — |
| MMLU-Pro | 55% | — |
| GPQA (HELM) | 50.7% | — |
Multimodal MiniMax-M3 leads
GPT-4.1 nano: 29.2 (#113), MiniMax-M3: 40.2 (#51)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1063 | 1253 |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
GPT-4.1 nano: 41.6 (#205), MiniMax-M3: 53.0 (#75)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1260 | 1420 |
| LMArena Chinese | 1270 | 1463 |
| LMArena German | 1288 | 1426 |
| LMArena Japanese | 1198 | 1381 |
| LMArena Russian | 1261 | 1428 |
| LMArena French | — | 1447 |
| LMArena Korean | — | 1372 |
| LMArena Spanish | — | 1432 |
Instruction Following MiniMax-M3 leads
GPT-4.1 nano: 67.8 (#193), MiniMax-M3: 75.5 (#62)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1267 | 1433 |
| IFEval | 84.3% | — |
Long Context MiniMax-M3 leads
GPT-4.1 nano: 23.7 (#296), MiniMax-M3: 44.2 (#72)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1283 | 1445 |
| Fiction.LiveBench | 25% | — |
Writing & Preference MiniMax-M3 leads
GPT-4.1 nano: 40.5 (#243), MiniMax-M3: 62.1 (#83)
| Benchmark | GPT-4.1 nano | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1285 | 1433 |
| LMArena Creative Writing | 1260 | 1404 |
| LMArena Multi-Turn | 1277 | 1442 |
| EQ-Bench Creative Writing | 946 | — |
| WildBench | 81.2% | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is GPT-4.1 nano better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.8 to 27.9 on the Noometry Index. GPT-4.1 nano costs 3.0× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 nano or MiniMax-M3?
GPT-4.1 nano is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is GPT-4.1 nano or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 24.1 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 nano does, with 1.05M tokens against 1M.
How many benchmarks do GPT-4.1 nano and MiniMax-M3 share?
22 benchmarks have published results for both models. GPT-4.1 nano has 38 scored results on Noometry and MiniMax-M3 has 41.