# gpt-oss-20b vs MiniMax-M2.1

> MiniMax-M2.1 is the stronger model overall, scoring 38.9 to 32.5 on the Noometry Index. gpt-oss-20b costs 15× less per token, which makes it the better buy when MiniMax-M2.1's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-oss-20b-vs-minimax-m2-1
- Last updated: 2026-10-12
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. gpt-oss-20b scores higher in 2 categories and MiniMax-M2.1 in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.1 leads 58.3 to 35.5.
- The biggest single-benchmark swing is Terminal-Bench: 3.4% for gpt-oss-20b and 36.6% for MiniMax-M2.1.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.1.
- MiniMax-M2.1 accepts more context: 205K tokens versus 131K.

## Snapshot

| | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 32.5 | 38.9 |
| Rank | 255 | 178 |
| Context | 131K | 205K |
| Input $/M | $0.018 | $0.30 |
| Output $/M | $0.09 | $1.20 |
| Weights | Open | Open |

## Coding

- gpt-oss-20b: 37.6 (#192)
- MiniMax-M2.1: 40.4 (#143)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| LMArena Coding | 1306 | 1421 |
| ALE-Bench | 566.05 | 623.83 |
| LMArena WebDev | — | 1384 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |

## Agentic & Tool Use

- gpt-oss-20b: 9.3 (#154)
- MiniMax-M2.1: 27.9 (#98)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| Terminal-Bench | 3.4% | 36.6% |

## Reasoning

- gpt-oss-20b: 19.3 (#261)
- MiniMax-M2.1: 16.6 (#302)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| LMArena Hard Prompts | 1274 | 1411 |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 11.2% |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.82 | — |

## Math

- gpt-oss-20b: 39.4 (#103)
- MiniMax-M2.1: 38.3 (#138)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| LMArena Math | 1317 | 1397 |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| Omni-MATH | 56.5% | — |

## Knowledge

- gpt-oss-20b: 34.6 (#195)
- MiniMax-M2.1: 38.3 (#147)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| LMArena Expert | 1258 | 1431 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| Vectara Hallucination Rate | — | 11.8% |
| GPQA (HELM) | 59.4% | — |

## Multilingual

- gpt-oss-20b: 42.2 (#197)
- MiniMax-M2.1: 50.0 (#128)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| LMArena Non-English | 1268 | 1378 |
| LMArena Chinese | 1314 | 1430 |
| LMArena German | 1255 | 1381 |
| LMArena Japanese | 1244 | 1287 |
| LMArena Korean | 1236 | 1298 |
| LMArena Russian | 1278 | 1387 |
| LMArena Spanish | 1267 | 1397 |
| LMArena French | — | 1404 |

## Instruction Following

- gpt-oss-20b: 61.8 (#240)
- MiniMax-M2.1: 73.8 (#112)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| LMArena Instruction Following | 1236 | 1400 |
| IFEval | 73.2% | — |

## Long Context

- gpt-oss-20b: 37.9 (#209)
- MiniMax-M2.1: 43.2 (#101)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| LMArena Longer Query | 1250 | 1416 |

## Writing & Preference

- gpt-oss-20b: 35.5 (#265)
- MiniMax-M2.1: 58.3 (#120)

| Benchmark | gpt-oss-20b | MiniMax-M2.1 |
|---|---|---|
| LMArena Text | 1287 | 1392 |
| LMArena Creative Writing | 1201 | 1361 |
| LMArena Multi-Turn | 1268 | 1396 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |

## FAQ

### Is gpt-oss-20b better than MiniMax-M2.1?

MiniMax-M2.1 is the stronger model overall, scoring 38.9 to 32.5 on the Noometry Index. gpt-oss-20b costs 15× less per token, which makes it the better buy when MiniMax-M2.1's lead doesn't matter for your workload.

### Which is cheaper, gpt-oss-20b or MiniMax-M2.1?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; MiniMax-M2.1 lists at $0.30 and $1.20.

### Is gpt-oss-20b or MiniMax-M2.1 better for coding?

MiniMax-M2.1 scores higher on coding benchmarks: 40.4 versus 37.6 in the Noometry coding category.

### Which has the bigger context window?

MiniMax-M2.1 does, with 205K tokens against 131K.

### How many benchmarks do gpt-oss-20b and MiniMax-M2.1 share?

18 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and MiniMax-M2.1 has 22.