# GPT-6.1 Sol vs MiniMax-M2.5

> GPT-6.1 Sol is the stronger model overall, scoring 65.6 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 7.6× less per token, which makes it the better buy when GPT-6.1 Sol's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-6-1-sol-vs-minimax-m2-5
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. GPT-6.1 Sol scores higher in 9 categories and MiniMax-M2.5 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6.1 Sol leads 93.7 to 26.9.
- The biggest single-benchmark swing is ProofBench: 99% for GPT-6.1 Sol and 4% for MiniMax-M2.5.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $2 / $10 for GPT-6.1 Sol.
- GPT-6.1 Sol accepts more context: 1.05M tokens versus 205K.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 65.6 | 38.3 |
| Rank | 6 | 188 |
| Context | 1.05M | 205K |
| Input $/M | $2 | $0.30 |
| Output $/M | $10 | $1.20 |
| Weights | Proprietary | Open |

## Coding

- GPT-6.1 Sol: 63.2 (#8)
- MiniMax-M2.5: 48.1 (#58)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| LMArena WebDev | 1755 | 1387 |
| LMArena Coding | 1487 | 1381 |
| DeepSWE | 75.2% | — |
| FrontierCode | 50.2% | — |
| SWE-bench Verified (bash only) | — | 75.8% |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 55.8% | — |
| ALE-Bench | — | 618.17 |

## Agentic & Tool Use

- GPT-6.1 Sol: 39.6 (#26)
- MiniMax-M2.5: 30.4 (#77)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| APEX-Agents | 60% | — |
| GDP.pdf | 32% | — |
| Vending-Bench 2 | — | -23.16 |

## Reasoning

- GPT-6.1 Sol: 81.9 (#2)
- MiniMax-M2.5: 17.5 (#292)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| ARC-AGI-2 | 94.2% | 4.9% |
| NYT Connections (extended) | 95.5% | 16.8% |
| ARC-AGI-1 | 98.5% | 63.7% |
| LMArena Hard Prompts | 1466 | 1372 |
| Epoch Capabilities Index | 166.09 | 146.68 |
| Kagi LLM Benchmark | — | 55.2% |
| CritPt | 31.7% | — |
| Chess Puzzles | 61% | — |
| EBR-Bench | 54.3% | — |
| Mystery Game Puzzles | 80% | — |

## Math

- GPT-6.1 Sol: 93.7 (#1)
- MiniMax-M2.5: 26.9 (#253)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| ProofBench | 99% | 4% |
| LMArena Math | 1464 | 1378 |
| FrontierMath (Tiers 1-3) | 93.7% | — |
| FrontierMath Tier 4 | 100% | — |
| OTIS Mock AIME 2024-2025 | 100% | — |

## Knowledge

- GPT-6.1 Sol: 71.8 (#4)
- MiniMax-M2.5: 39.2 (#135)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| LMArena Expert | 1502 | 1379 |
| GPQA Diamond | 95.4% | — |
| SimpleQA Verified | 73.9% | — |
| Vectara Hallucination Rate | — | 9.1% |

## Multimodal

- GPT-6.1 Sol: 52.7 (#5)
- MiniMax-M2.5: —

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| LMArena Vision | 1288 | — |
| Furniture Assembly | 80% | — |

## Multilingual

- GPT-6.1 Sol: 54.3 (#46)
- MiniMax-M2.5: 47.1 (#152)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1438 | 1338 |
| LMArena Chinese | 1477 | 1393 |
| LMArena Russian | 1455 | 1358 |
| LMArena French | — | 1362 |
| LMArena German | — | 1362 |
| LMArena Japanese | — | 1171 |
| LMArena Korean | — | 1232 |
| LMArena Spanish | — | 1354 |

## Instruction Following

- GPT-6.1 Sol: 77.0 (#29)
- MiniMax-M2.5: 71.5 (#148)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1468 | 1353 |

## Long Context

- GPT-6.1 Sol: 44.9 (#54)
- MiniMax-M2.5: 37.5 (#216)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1465 | 1366 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |

## Writing & Preference

- GPT-6.1 Sol: 63.6 (#63)
- MiniMax-M2.5: 53.9 (#153)

| Benchmark | GPT-6.1 Sol | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1447 | 1359 |
| LMArena Creative Writing | 1432 | 1331 |
| LMArena Multi-Turn | 1449 | 1364 |
| EQ-Bench Creative Writing | — | 1361 |

## FAQ

### Is GPT-6.1 Sol better than MiniMax-M2.5?

GPT-6.1 Sol is the stronger model overall, scoring 65.6 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 7.6× less per token, which makes it the better buy when GPT-6.1 Sol's lead doesn't matter for your workload.

### Which is cheaper, GPT-6.1 Sol or MiniMax-M2.5?

MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-6.1 Sol lists at $2 and $10.

### Is GPT-6.1 Sol or MiniMax-M2.5 better for coding?

GPT-6.1 Sol scores higher on coding benchmarks: 63.2 versus 48.1 in the Noometry coding category.

### Which has the bigger context window?

GPT-6.1 Sol does, with 1.05M tokens against 205K.

### How many benchmarks do GPT-6.1 Sol and MiniMax-M2.5 share?

18 benchmarks have published results for both models. GPT-6.1 Sol has 34 scored results on Noometry and MiniMax-M2.5 has 33.