# gpt-oss-20b vs MiniMax-M2.5

> MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 32.5 on the Noometry Index. gpt-oss-20b costs 15× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-oss-20b-vs-minimax-m2-5
- Last updated: 2026-10-12
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. gpt-oss-20b scores higher in 3 categories and MiniMax-M2.5 in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where MiniMax-M2.5 leads 30.4 to 9.3.
- The biggest single-benchmark swing is Terminal-Bench: 3.4% for gpt-oss-20b and 42.7% for MiniMax-M2.5.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
- MiniMax-M2.5 accepts more context: 205K tokens versus 131K.

## Snapshot

| | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 32.5 | 38.3 |
| Rank | 255 | 188 |
| 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-M2.5: 48.1 (#58)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| LMArena Coding | 1306 | 1381 |
| ALE-Bench | 566.05 | 618.17 |
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1387 |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |

## Agentic & Tool Use

- gpt-oss-20b: 9.3 (#154)
- MiniMax-M2.5: 30.4 (#77)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | 3.4% | 42.7% |
| Vending-Bench 2 | — | -23.16 |

## Reasoning

- gpt-oss-20b: 19.3 (#261)
- MiniMax-M2.5: 17.5 (#292)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| Kagi LLM Benchmark | 53.2% | 55.2% |
| LMArena Hard Prompts | 1274 | 1372 |
| Epoch Capabilities Index | 137.82 | 146.68 |
| ARC-AGI-2 | — | 4.9% |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |

## Math

- gpt-oss-20b: 39.4 (#103)
- MiniMax-M2.5: 26.9 (#253)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1317 | 1378 |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| ProofBench | — | 4% |
| Omni-MATH | 56.5% | — |

## Knowledge

- gpt-oss-20b: 34.6 (#195)
- MiniMax-M2.5: 39.2 (#135)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| LMArena Expert | 1258 | 1379 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| Vectara Hallucination Rate | — | 9.1% |
| GPQA (HELM) | 59.4% | — |

## Multilingual

- gpt-oss-20b: 42.2 (#197)
- MiniMax-M2.5: 47.1 (#152)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1268 | 1338 |
| LMArena Chinese | 1314 | 1393 |
| LMArena German | 1255 | 1362 |
| LMArena Japanese | 1244 | 1171 |
| LMArena Korean | 1236 | 1232 |
| LMArena Russian | 1278 | 1358 |
| LMArena Spanish | 1267 | 1354 |
| LMArena French | — | 1362 |

## Instruction Following

- gpt-oss-20b: 61.8 (#240)
- MiniMax-M2.5: 71.5 (#148)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1236 | 1353 |
| IFEval | 73.2% | — |

## Long Context

- gpt-oss-20b: 37.9 (#209)
- MiniMax-M2.5: 37.5 (#216)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1250 | 1366 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |

## Writing & Preference

- gpt-oss-20b: 35.5 (#265)
- MiniMax-M2.5: 53.9 (#153)

| Benchmark | gpt-oss-20b | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1287 | 1359 |
| LMArena Creative Writing | 1201 | 1331 |
| EQ-Bench Creative Writing | 666 | 1361 |
| LMArena Multi-Turn | 1268 | 1364 |
| WildBench | 73.7% | — |

## FAQ

### Is gpt-oss-20b better than MiniMax-M2.5?

MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 32.5 on the Noometry Index. gpt-oss-20b costs 15× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.

### Which is cheaper, gpt-oss-20b or MiniMax-M2.5?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.

### Is gpt-oss-20b or MiniMax-M2.5 better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 37.6 in the Noometry coding category.

### Which has the bigger context window?

MiniMax-M2.5 does, with 205K tokens against 131K.

### How many benchmarks do gpt-oss-20b and MiniMax-M2.5 share?

21 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and MiniMax-M2.5 has 33.