Model comparison
gpt-oss-20b vs Granite 4.2 3b
Granite 4.2 3b is the stronger model overall, scoring 39.0 to 32.1 on the Noometry Index.
Last verified . 11 shared benchmarks.
Summary
- They share 11 benchmarks with published results for both. gpt-oss-20b scores higher in 1 category and Granite 4.2 3b in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Granite 4.2 3b leads 47.9 to 35.8.
Side by side
| gpt-oss-20b | Granite 4.2 3b | |
|---|---|---|
| Provider | OpenAI | IBM |
| Noometry Index | 32.1 | 39.0 |
| Released | 2025-08-05 | — |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.018 | — |
| Output $ / M tokens | $0.09 | — |
| Results tracked | 34 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Granite 4.2 3b leads
gpt-oss-20b: 37.1 (#185), Granite 4.2 3b: 39.4 (#146)
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| LMArena Coding | 1307 | 1363 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| ALE-Bench | 566.05 | — |
Agentic & Tool Use Not comparable
gpt-oss-20b: 9.2 (#146), Granite 4.2 3b: —
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| Terminal-Bench | 3.4% | — |
Reasoning Granite 4.2 3b leads
gpt-oss-20b: 18.8 (#251), Granite 4.2 3b: 25.0 (#136)
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| LMArena Hard Prompts | 1274 | 1308 |
| Kagi LLM Benchmark | 53.2% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.81 | — |
Math Not comparable
gpt-oss-20b: 37.6 (#93), Granite 4.2 3b: —
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| Omni-MATH | 56.5% | — |
| LMArena Math | 1317 | — |
Knowledge Granite 4.2 3b leads
gpt-oss-20b: 34.6 (#186), Granite 4.2 3b: 35.6 (#169)
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| LMArena Expert | 1258 | 1313 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
Multilingual Too close to call
gpt-oss-20b: 42.4 (#192), Granite 4.2 3b: 42.4 (#193)
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| LMArena Non-English | 1269 | 1268 |
| LMArena Chinese | 1311 | 1268 |
| LMArena Russian | 1279 | 1249 |
| LMArena German | 1249 | — |
| LMArena Japanese | 1245 | — |
| LMArena Korean | 1239 | — |
| LMArena Spanish | 1267 | — |
Instruction Following Granite 4.2 3b leads
gpt-oss-20b: 62.0 (#235), Granite 4.2 3b: 67.2 (#195)
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| LMArena Instruction Following | 1236 | 1271 |
| IFEval | 73.2% | — |
Long Context Granite 4.2 3b leads
gpt-oss-20b: 37.9 (#203), Granite 4.2 3b: 39.3 (#179)
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| LMArena Longer Query | 1251 | 1293 |
Writing & Preference Granite 4.2 3b leads
gpt-oss-20b: 35.8 (#260), Granite 4.2 3b: 47.9 (#202)
| Benchmark | gpt-oss-20b | Granite 4.2 3b |
|---|---|---|
| LMArena Text | 1288 | 1295 |
| LMArena Creative Writing | 1201 | 1215 |
| LMArena Multi-Turn | 1268 | 1293 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
Frequently asked questions
Is gpt-oss-20b better than Granite 4.2 3b?
Granite 4.2 3b is the stronger model overall, scoring 39.0 to 32.1 on the Noometry Index.
Is gpt-oss-20b or Granite 4.2 3b better for coding?
Granite 4.2 3b scores higher on coding benchmarks: 39.4 versus 37.1 in the Noometry coding category.
How many benchmarks do gpt-oss-20b and Granite 4.2 3b share?
11 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Granite 4.2 3b has 11.