Model comparison
GPT-5 Nano vs MiniMax-M2
MiniMax-M2 is the stronger model overall, scoring 37.4 to 33.5 on the Noometry Index. GPT-5 Nano costs 3.8× less per token, which makes it the better buy when MiniMax-M2's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
MiniMaxSummary
- They share 17 benchmarks with published results for both. GPT-5 Nano scores higher in 2 categories and MiniMax-M2 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2 leads 53.0 to 39.1.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 34.8% for GPT-5 Nano and 61% for MiniMax-M2.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.
- GPT-5 Nano accepts more context: 400K tokens versus 205K.
- MiniMax-M2 has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Nano | MiniMax-M2 | |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 33.5 | 37.4 |
| Released | 2025-08-07 | 2025-10-27 |
| Weights | Proprietary | Open |
| Context window | 400K | 205K |
| Max output | 128K | 131K |
| Input $ / M tokens | $0.05 | $0.30 |
| Output $ / M tokens | $0.40 | $1.20 |
| Results tracked | 49 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2 leads
GPT-5 Nano: 33.6 (#254), MiniMax-M2: 39.3 (#159)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| SWE-bench Verified (bash only) | 34.8% | 61% |
| LMArena Coding | 1351 | 1370 |
| LMArena WebDev | — | 1297 |
| WeirdML | 38.1% | — |
| ALE-Bench | 718.67 | — |
Agentic & Tool Use Too close to call
GPT-5 Nano: 25.8 (#106), MiniMax-M2: 25.1 (#109)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| Terminal-Bench | 21.8% | 30% |
| Berkeley Function Calling Leaderboard | 51.5% | — |
| Vending-Bench 2 | — | 160.6 |
Reasoning MiniMax-M2 leads
GPT-5 Nano: 16.3 (#306), MiniMax-M2: 19.4 (#258)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| Kagi LLM Benchmark | 62.2% | 57.8% |
| LMArena Hard Prompts | 1328 | 1357 |
| ARC-AGI-2 | 2.6% | — |
| NYT Connections (extended) | — | 14.8% |
| ARC-AGI-1 | 20.7% | — |
| Chess Puzzles | 27% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 62.7% | — |
| LMCA | 7.9% | — |
| Epoch Capabilities Index | 139.38 | — |
| ForecastBench | 59.1 | — |
Math MiniMax-M2 leads
GPT-5 Nano: 29.4 (#241), MiniMax-M2: 37.3 (#160)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| LMArena Math | 1317 | 1352 |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| OTIS Mock AIME 2024-2025 | 81.1% | — |
| ProofBench | 12% | — |
| Omni-MATH | 54.6% | — |
| MATH Level 5 | 95.2% | — |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge MiniMax-M2 leads
GPT-5 Nano: 35.9 (#178), MiniMax-M2: 37.0 (#163)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| LMArena Expert | 1321 | 1337 |
| GPQA Diamond | 69.4% | — |
| SimpleQA Verified | 11.7% | — |
| MMLU-Pro | 77.8% | — |
| Vectara Hallucination Rate | 10.5% | — |
| GPQA (HELM) | 67.9% | — |
Multimodal Not comparable
GPT-5 Nano: 31.3 (#108), MiniMax-M2: —
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| LMArena Vision | 1159 | — |
| VPCT | 37.2% | — |
Multilingual Too close to call
GPT-5 Nano: 45.3 (#172), MiniMax-M2: 45.3 (#171)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| LMArena Non-English | 1313 | 1313 |
| LMArena Chinese | 1356 | 1366 |
| LMArena German | 1327 | 1355 |
| LMArena Russian | 1296 | 1331 |
| LMArena Spanish | 1360 | 1326 |
| LMArena French | — | 1335 |
| LMArena Japanese | 1226 | — |
| LMArena Korean | 1269 | — |
Instruction Following GPT-5 Nano leads
GPT-5 Nano: 75.0 (#79), MiniMax-M2: 70.2 (#166)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| LMArena Instruction Following | 1306 | 1328 |
| IFEval | 93.2% | — |
Long Context MiniMax-M2 leads
GPT-5 Nano: 31.3 (#281), MiniMax-M2: 40.5 (#153)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| LMArena Longer Query | 1312 | 1331 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference MiniMax-M2 leads
GPT-5 Nano: 39.1 (#249), MiniMax-M2: 53.0 (#162)
| Benchmark | GPT-5 Nano | MiniMax-M2 |
|---|---|---|
| LMArena Text | 1320 | 1340 |
| LMArena Creative Writing | 1249 | 1286 |
| LMArena Multi-Turn | 1311 | 1361 |
| EQ-Bench Creative Writing | 705 | — |
| WildBench | 80.6% | — |
Frequently asked questions
Is GPT-5 Nano better than MiniMax-M2?
MiniMax-M2 is the stronger model overall, scoring 37.4 to 33.5 on the Noometry Index. GPT-5 Nano costs 3.8× less per token, which makes it the better buy when MiniMax-M2's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Nano or MiniMax-M2?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; MiniMax-M2 lists at $0.30 and $1.20.
Is GPT-5 Nano or MiniMax-M2 better for coding?
MiniMax-M2 scores higher on coding benchmarks: 39.3 versus 33.6 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Nano does, with 400K tokens against 205K.
How many benchmarks do GPT-5 Nano and MiniMax-M2 share?
17 benchmarks have published results for both models. GPT-5 Nano has 49 scored results on Noometry and MiniMax-M2 has 21.