Model comparison
GPT-5.4 nano vs MiniMax-M2.7
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 37.7 on the Noometry Index.
Last verified . 24 shared benchmarks.
MiniMaxSummary
- They share 24 benchmarks with published results for both. GPT-5.4 nano scores higher in 4 categories and MiniMax-M2.7 in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 nano leads 40.9 to 25.9.
- The biggest single-benchmark swing is WeirdML: 49.2% for GPT-5.4 nano and 37% for MiniMax-M2.7.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.7.
- GPT-5.4 nano accepts more context: 400K tokens versus 205K.
- MiniMax-M2.7 has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 nano | MiniMax-M2.7 | |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 41.9 | 37.7 |
| Released | 2026-03-17 | 2026-03-18 |
| Weights | Proprietary | Open |
| Context window | 400K | 205K |
| Max output | 128K | 131K |
| Input $ / M tokens | $0.20 | $0.30 |
| Output $ / M tokens | $1.25 | $1.20 |
| Results tracked | 40 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GPT-5.4 nano: 43.6 (#84), MiniMax-M2.7: 41.8 (#120)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| SciCode | 46.9% | 47% |
| WeirdML | 49.2% | 37% |
| LMArena Coding | 1405 | 1454 |
| ALE-Bench | 1,005 | 599.25 |
| LMArena WebDev | — | 1398 |
Agentic & Tool Use Not comparable
GPT-5.4 nano: —, MiniMax-M2.7: 25.1 (#111)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| Terminal-Bench | — | 45.1% |
| ExploitBench | — | 13.3% |
| GBAEval | — | 0% |
Reasoning GPT-5.4 nano leads
GPT-5.4 nano: 23.7 (#173), MiniMax-M2.7: 19.7 (#253)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| CritPt | 9.3% | 0.6% |
| LMArena Hard Prompts | 1381 | 1422 |
| Epoch Capabilities Index | 145.81 | 145.85 |
| ARC-AGI-2 | 5.7% | — |
| Kagi LLM Benchmark | 39.7% | — |
| NYT Connections (extended) | — | 24.7% |
| ARC-AGI-1 | 51.5% | — |
| Chess Puzzles | 30% | — |
| Thematic Generalization | — | 39.3% |
| Mystery Game Puzzles | 9% | — |
| DTBench | 80.3% | — |
| LMCA | 36.9% | — |
| ForecastBench | 57.3 | — |
Math GPT-5.4 nano leads
GPT-5.4 nano: 40.9 (#88), MiniMax-M2.7: 25.9 (#263)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| ProofBench | 5% | 3% |
| LMArena Math | 1406 | 1420 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5.4 nano leads
GPT-5.4 nano: 41.9 (#103), MiniMax-M2.7: 37.7 (#152)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| Vectara Hallucination Rate | 3.1% | 12.9% |
| LMArena Expert | 1396 | 1444 |
| GPQA Diamond | 78.5% | — |
| SimpleQA Verified | 11.7% | — |
Multimodal Not comparable
GPT-5.4 nano: 36.7 (#78), MiniMax-M2.7: —
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| LMArena Vision | 1196 | — |
Multilingual MiniMax-M2.7 leads
GPT-5.4 nano: 48.6 (#140), MiniMax-M2.7: 50.3 (#123)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| LMArena Non-English | 1359 | 1382 |
| LMArena Chinese | 1392 | 1441 |
| LMArena French | 1396 | 1421 |
| LMArena German | 1367 | 1398 |
| LMArena Japanese | 1343 | 1262 |
| LMArena Korean | 1320 | 1313 |
| LMArena Russian | 1363 | 1383 |
| LMArena Spanish | 1371 | 1403 |
Instruction Following MiniMax-M2.7 leads
GPT-5.4 nano: 71.9 (#144), MiniMax-M2.7: 74.1 (#103)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| LMArena Instruction Following | 1362 | 1405 |
Long Context MiniMax-M2.7 leads
GPT-5.4 nano: 41.6 (#137), MiniMax-M2.7: 43.3 (#99)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| LMArena Longer Query | 1366 | 1419 |
Writing & Preference MiniMax-M2.7 leads
GPT-5.4 nano: 55.7 (#142), MiniMax-M2.7: 58.9 (#112)
| Benchmark | GPT-5.4 nano | MiniMax-M2.7 |
|---|---|---|
| LMArena Text | 1372 | 1405 |
| LMArena Creative Writing | 1314 | 1354 |
| LMArena Multi-Turn | 1382 | 1412 |
Frequently asked questions
Is GPT-5.4 nano better than MiniMax-M2.7?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 37.7 on the Noometry Index.
Which is cheaper, GPT-5.4 nano or MiniMax-M2.7?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; MiniMax-M2.7 lists at $0.30 and $1.20.
Is GPT-5.4 nano or MiniMax-M2.7 better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 205K.
How many benchmarks do GPT-5.4 nano and MiniMax-M2.7 share?
24 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and MiniMax-M2.7 has 30.