# gpt-oss-20b vs Granite 4.0 Micro

> gpt-oss-20b is the stronger model overall, scoring 32.1 to 28.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-oss-20b-vs-granite-4-0-micro
- Last updated: 2026-10-10
- Shared benchmarks: 8

## Summary

- They share 8 benchmarks with published results for both. gpt-oss-20b scores higher in 3 categories and Granite 4.0 Micro in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-20b leads 37.6 to 11.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 65.3% for gpt-oss-20b and 2.8% for Granite 4.0 Micro.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.017 / $0.11 for Granite 4.0 Micro.
- gpt-oss-20b accepts more context: 131K tokens versus 131K.

## Snapshot

| | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| Provider | OpenAI | IBM |
| Noometry Index | 32.1 | 28.8 |
| Rank | 249 | 312 |
| Context | 131K | 131K |
| Input $/M | $0.018 | $0.017 |
| Output $/M | $0.09 | $0.11 |
| Weights | Open | Open |

## Coding

- gpt-oss-20b: 37.1 (#185)
- Granite 4.0 Micro: —

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| LMArena Coding | 1307 | — |
| ALE-Bench | 566.05 | — |

## Agentic & Tool Use

- gpt-oss-20b: 9.2 (#146)
- Granite 4.0 Micro: —

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| Terminal-Bench | 3.4% | — |

## Reasoning

- gpt-oss-20b: 18.8 (#251)
- Granite 4.0 Micro: 18.5 (#261)

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| Chess Puzzles | 4% | 0% |
| Kagi LLM Benchmark | 53.2% | — |
| CritPt | 1.4% | — |
| LMArena Hard Prompts | 1274 | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.81 | — |

## Math

- gpt-oss-20b: 37.6 (#93)
- Granite 4.0 Micro: 11.2 (#301)

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | 2.8% |
| Omni-MATH | 56.5% | 20.9% |
| LMArena Math | 1317 | — |

## Knowledge

- gpt-oss-20b: 34.6 (#186)
- Granite 4.0 Micro: 9.9 (#299)

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| GPQA Diamond | 60.8% | 28.3% |
| MMLU-Pro | 74% | 39.5% |
| GPQA (HELM) | 59.4% | 30.7% |
| LMArena Expert | 1258 | — |

## Multilingual

- gpt-oss-20b: 42.4 (#192)
- Granite 4.0 Micro: —

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| LMArena Non-English | 1269 | — |
| LMArena Chinese | 1311 | — |
| LMArena German | 1249 | — |
| LMArena Japanese | 1245 | — |
| LMArena Korean | 1239 | — |
| LMArena Russian | 1279 | — |
| LMArena Spanish | 1267 | — |

## Instruction Following

- gpt-oss-20b: 62.0 (#235)
- Granite 4.0 Micro: 70.0 (#164)

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| IFEval | 73.2% | 84.9% |
| LMArena Instruction Following | 1236 | — |

## Long Context

- gpt-oss-20b: 37.9 (#203)
- Granite 4.0 Micro: —

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| LMArena Longer Query | 1251 | — |

## Writing & Preference

- gpt-oss-20b: 35.8 (#260)
- Granite 4.0 Micro: 46.7 (#212)

| Benchmark | gpt-oss-20b | Granite 4.0 Micro |
|---|---|---|
| WildBench | 73.7% | 67% |
| LMArena Text | 1288 | — |
| LMArena Creative Writing | 1201 | — |
| EQ-Bench Creative Writing | 666 | — |
| LMArena Multi-Turn | 1268 | — |

## FAQ

### Is gpt-oss-20b better than Granite 4.0 Micro?

gpt-oss-20b is the stronger model overall, scoring 32.1 to 28.8 on the Noometry Index.

### Which is cheaper, gpt-oss-20b or Granite 4.0 Micro?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Granite 4.0 Micro lists at $0.017 and $0.11.

### Which has the bigger context window?

gpt-oss-20b does, with 131K tokens against 131K.

### How many benchmarks do gpt-oss-20b and Granite 4.0 Micro share?

8 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Granite 4.0 Micro has 8.