Model comparison
GPT-4o mini vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.5 to 25.4 on the Noometry Index. GPT-4o mini costs 2.0× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
MiniMaxSummary
- They share 26 benchmarks with published results for both. GPT-4o mini 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 17.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.9% for GPT-4o mini and 71.1% for MiniMax-M3.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 accepts more context: 1M tokens versus 128K.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| GPT-4o mini | MiniMax-M3 | |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 25.4 | 43.5 |
| Released | 2024-07-18 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | 128K | 1M |
| Max output | 16K | 512K |
| Input $ / M tokens | $0.15 | $0.30 |
| Output $ / M tokens | $0.60 | $1.20 |
| Results tracked | 60 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
GPT-4o mini: 21.6 (#330), MiniMax-M3: 41.5 (#111)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| LMArena Coding | 1290 | 1470 |
| FrontierCode | — | 14.7% |
| Aider Polyglot | 3.6% | — |
| LMArena WebDev | — | 1482 |
| SciCode | — | 47.1% |
| WeirdML | 11.8% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 43.1% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 640.02 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use GPT-4o mini leads
GPT-4o mini: 27.2 (#95), MiniMax-M3: 22.5 (#121)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| BALROG | 17.4% | — |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
GPT-4o mini: 8.7 (#343), MiniMax-M3: 29.3 (#83)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| SimpleBench | 10.7% | 45.8% |
| Chess Puzzles | 0% | 14% |
| LMArena Hard Prompts | 1267 | 1447 |
| Mystery Game Puzzles | 12% | 8% |
| DTBench | 54.4% | 78.9% |
| LMCA | 10.4% | 33.7% |
| Epoch Capabilities Index | 126.56 | 147 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 28.8% | — |
| NYT Connections (extended) | — | 65.1% |
| CritPt | — | 3.7% |
| LiveBench Reasoning | 32.8% | — |
| LiveBench Data Analysis | 50% | — |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
| LiveBench | 41.3% | — |
| PIQA | 88.7% | — |
Math MiniMax-M3 leads
GPT-4o mini: 9.9 (#307), MiniMax-M3: 39.1 (#83)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.9% | 71.1% |
| LMArena Math | 1267 | 1432 |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| ProofBench | — | 18% |
| Omni-MATH | 28% | — |
| LiveBench Math | 36.3% | — |
| MATH Level 5 | 52.6% | — |
| GSM8K | 91.3% | — |
Knowledge MiniMax-M3 leads
GPT-4o mini: 17.7 (#279), MiniMax-M3: 58.4 (#32)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 37.7% | 90.9% |
| LMArena Expert | 1235 | 1462 |
| SimpleQA Verified | 8.3% | — |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| GPQA (HELM) | 36.8% | — |
| BoolQ | 88.7% | — |
| MMLU | 81.8% | — |
Multimodal MiniMax-M3 leads
GPT-4o mini: 25.3 (#117), MiniMax-M3: 39.5 (#45)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1066 | 1254 |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
GPT-4o mini: 42.2 (#194), MiniMax-M3: 53.2 (#70)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1266 | 1420 |
| LMArena Chinese | 1264 | 1465 |
| LMArena French | 1296 | 1445 |
| LMArena German | 1272 | 1424 |
| LMArena Japanese | 1216 | 1381 |
| LMArena Korean | 1195 | 1372 |
| LMArena Russian | 1275 | 1430 |
| LMArena Spanish | 1276 | 1431 |
Instruction Following MiniMax-M3 leads
GPT-4o mini: 62.1 (#234), MiniMax-M3: 75.7 (#56)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1259 | 1433 |
| LiveBench Instruction Following | 56.8% | — |
| IFEval | 78.2% | — |
Long Context MiniMax-M3 leads
GPT-4o mini: 39.1 (#181), MiniMax-M3: 44.2 (#68)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1289 | 1445 |
Writing & Preference MiniMax-M3 leads
GPT-4o mini: 39.7 (#243), MiniMax-M3: 62.3 (#79)
| Benchmark | GPT-4o mini | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1286 | 1433 |
| LMArena Creative Writing | 1268 | 1405 |
| LMArena Multi-Turn | 1285 | 1441 |
| Short-Story Creative Writing | 67.2% | — |
| EQ-Bench Creative Writing | 873 | — |
| WildBench | 79.1% | — |
| EQ-Bench 4 | — | 1150 |
| LiveBench Language | 28.6% | — |
Frequently asked questions
Is GPT-4o mini better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.5 to 25.4 on the Noometry Index. GPT-4o mini costs 2.0× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.
Which is cheaper, GPT-4o mini or MiniMax-M3?
GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is GPT-4o mini or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 21.6 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 128K.
How many benchmarks do GPT-4o mini and MiniMax-M3 share?
26 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and MiniMax-M3 has 41.