# GPT-5.5 vs MiniMax-M3

> GPT-5.5 is the stronger model overall, scoring 63.4 to 43.5 on the Noometry Index. MiniMax-M3 costs 21× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-5-vs-minimax-m3
- Last updated: 2026-10-10
- Shared benchmarks: 41

## Summary

- They share 41 benchmarks with published results for both. GPT-5.5 scores higher in 10 categories and MiniMax-M3 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.5 leads 72.7 to 29.3.
- The biggest single-benchmark swing is GBAEval: 53.2% for GPT-5.5 and 0.9% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 1M.
- MiniMax-M3 has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 63.4 | 43.5 |
| Rank | 8 | 81 |
| Context | 1.05M | 1M |
| Input $/M | $5 | $0.30 |
| Output $/M | $30 | $1.20 |
| Weights | Proprietary | Open |

## Coding

- GPT-5.5: 58.3 (#13)
- MiniMax-M3: 41.5 (#111)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| FrontierCode | 43% | 14.7% |
| LMArena WebDev | 1512 | 1482 |
| SciCode | 56.1% | 47.1% |
| LMArena Coding | 1494 | 1470 |
| ALE-Bench | 1,943 | 640.02 |
| SWE-bench Verified | 80.6% | — |
| DeepSWE | 67% | — |
| GSO | 40.2% | — |
| WeirdML | 84.9% | — |
| MirrorCode | 10% | — |

## Agentic & Tool Use

- GPT-5.5: 50.7 (#6)
- MiniMax-M3: 22.5 (#121)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| APEX-Agents | 55.1% | 37.7% |
| OSWorld 2.0 | 13% | 4.6% |
| GBAEval | 53.2% | 0.9% |
| Vending-Bench 2 | 7,524 | 2,158 |
| Terminal-Bench | 84.7% | — |
| Remote Labor Index | 6.3% | — |
| τ²-bench Banking | 44.6% | — |
| DeepResearch Bench | 54% | — |
| PostTrainBench | 27.2% | — |
| ExploitBench | 47.4% | — |
| GDP.pdf | 26% | — |
| LMArena Search | 1242 | — |

## Reasoning

- GPT-5.5: 72.7 (#8)
- MiniMax-M3: 29.3 (#83)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| SimpleBench | 69% | 45.8% |
| NYT Connections (extended) | 96.2% | 65.1% |
| CritPt | 27.1% | 3.7% |
| Chess Puzzles | 54% | 14% |
| LMArena Hard Prompts | 1489 | 1447 |
| Mystery Game Puzzles | 56% | 8% |
| DTBench | 96% | 78.9% |
| LMCA | 54.3% | 33.7% |
| Surface Evolver Bench | 88.1% | 55% |
| Epoch Capabilities Index | 159.25 | 147 |
| ForecastBench | 60.6 | 61.4 |
| ARC-AGI-2 | 85% | — |
| Kagi LLM Benchmark | 88.8% | — |
| ARC-AGI-1 | 95% | — |
| EBR-Bench | 34.3% | — |
| Bench to the Future 3 | 0.14 | — |

## Math

- GPT-5.5: 81.5 (#9)
- MiniMax-M3: 39.1 (#83)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 71.1% |
| ProofBench | 50% | 18% |
| LMArena Math | 1486 | 1432 |
| FrontierMath (Tiers 1-3) | 85.3% | — |
| FrontierMath Tier 4 | 72.5% | — |
| MathArena Final-Answer Competitions | 94.3% | — |
| FrontierMath (Feb 2025 set) | 51.7% | — |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | 35.4% | — |

## Knowledge

- GPT-5.5: 64.4 (#14)
- MiniMax-M3: 58.4 (#32)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 94% | 90.9% |
| LMArena Expert | 1508 | 1462 |
| SimpleQA Verified | 63% | — |
| Vectara Hallucination Rate | 9.3% | — |

## Multimodal

- GPT-5.5: 47.1 (#11)
- MiniMax-M3: 39.5 (#45)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1297 | 1254 |
| LMArena Document | 1486 | 1435 |
| Blueprint-Bench 2 | 36.2% | — |
| Furniture Assembly | 44.2% | — |

## Multilingual

- GPT-5.5: 56.5 (#19)
- MiniMax-M3: 53.2 (#70)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1467 | 1420 |
| LMArena Chinese | 1535 | 1465 |
| LMArena French | 1486 | 1445 |
| LMArena German | 1478 | 1424 |
| LMArena Japanese | 1498 | 1381 |
| LMArena Korean | 1461 | 1372 |
| LMArena Russian | 1473 | 1430 |
| LMArena Spanish | 1469 | 1431 |

## Instruction Following

- GPT-5.5: 77.7 (#18)
- MiniMax-M3: 75.7 (#56)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1479 | 1433 |

## Long Context

- GPT-5.5: 48.3 (#12)
- MiniMax-M3: 44.2 (#68)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1484 | 1445 |
| CL-bench Life | 22.2% | — |

## Writing & Preference

- GPT-5.5: 72.8 (#13)
- MiniMax-M3: 62.3 (#79)

| Benchmark | GPT-5.5 | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1471 | 1433 |
| LMArena Creative Writing | 1455 | 1405 |
| EQ-Bench 4 | 1315 | 1150 |
| LMArena Multi-Turn | 1477 | 1441 |
| EQ-Bench Creative Writing | 1844 | — |

## FAQ

### Is GPT-5.5 better than MiniMax-M3?

GPT-5.5 is the stronger model overall, scoring 63.4 to 43.5 on the Noometry Index. MiniMax-M3 costs 21× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

### Which is cheaper, GPT-5.5 or MiniMax-M3?

MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-5.5 lists at $5 and $30.

### Is GPT-5.5 or MiniMax-M3 better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.3 versus 41.5 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.5 does, with 1.05M tokens against 1M.

### How many benchmarks do GPT-5.5 and MiniMax-M3 share?

41 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and MiniMax-M3 has 41.