Model comparison
gpt-oss-20b vs Llama 3.2 90B
gpt-oss-20b is the stronger model overall, scoring 32.5 to 27.5 on the Noometry Index.
Last verified . 3 shared benchmarks.
MetaSummary
- They share 3 benchmarks with published results for both. gpt-oss-20b scores higher in 2 categories and Llama 3.2 90B in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-20b leads 39.4 to 11.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 65.3% for gpt-oss-20b and 2.6% for Llama 3.2 90B.
Side by side
| gpt-oss-20b | Llama 3.2 90B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 32.5 | 27.5 |
| Released | 2025-08-05 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.018 | — |
| Output $ / M tokens | $0.09 | — |
| Results tracked | 34 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
gpt-oss-20b: 37.6 (#192), Llama 3.2 90B: —
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| LMArena Coding | 1306 | — |
| ALE-Bench | 566.05 | — |
Agentic & Tool Use Llama 3.2 90B leads
gpt-oss-20b: 9.3 (#154), Llama 3.2 90B: 30.0 (#80)
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| Terminal-Bench | 3.4% | — |
| BALROG | — | 27.3% |
Reasoning Llama 3.2 90B leads
gpt-oss-20b: 19.3 (#261), Llama 3.2 90B: 21.7 (#217)
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| Epoch Capabilities Index | 137.82 | 125.5 |
| Kagi LLM Benchmark | 53.2% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| EnigmaEval | — | 0.4% |
| LMArena Hard Prompts | 1274 | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
Math gpt-oss-20b leads
gpt-oss-20b: 39.4 (#103), Llama 3.2 90B: 11.1 (#308)
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | 2.6% |
| Omni-MATH | 56.5% | — |
| LMArena Math | 1317 | — |
| MATH Level 5 | — | 39.4% |
Knowledge gpt-oss-20b leads
gpt-oss-20b: 34.6 (#195), Llama 3.2 90B: 21.7 (#274)
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| GPQA Diamond | 60.8% | 41% |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
| LMArena Expert | 1258 | — |
| MMLU | — | 80.3% |
Multimodal Not comparable
gpt-oss-20b: —, Llama 3.2 90B: 25.4 (#124)
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| LMArena Vision | — | 1000 |
| GeoBench | — | 52% |
Multilingual Not comparable
gpt-oss-20b: 42.2 (#197), Llama 3.2 90B: —
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| LMArena Non-English | 1268 | — |
| LMArena Chinese | 1314 | — |
| LMArena German | 1255 | — |
| LMArena Japanese | 1244 | — |
| LMArena Korean | 1236 | — |
| LMArena Russian | 1278 | — |
| LMArena Spanish | 1267 | — |
Instruction Following Not comparable
gpt-oss-20b: 61.8 (#240), Llama 3.2 90B: —
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| IFEval | 73.2% | — |
| LMArena Instruction Following | 1236 | — |
Long Context Not comparable
gpt-oss-20b: 37.9 (#209), Llama 3.2 90B: —
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| LMArena Longer Query | 1250 | — |
Writing & Preference Not comparable
gpt-oss-20b: 35.5 (#265), Llama 3.2 90B: —
| Benchmark | gpt-oss-20b | Llama 3.2 90B |
|---|---|---|
| LMArena Text | 1287 | — |
| LMArena Creative Writing | 1201 | — |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
| LMArena Multi-Turn | 1268 | — |
Frequently asked questions
Is gpt-oss-20b better than Llama 3.2 90B?
gpt-oss-20b is the stronger model overall, scoring 32.5 to 27.5 on the Noometry Index.
How many benchmarks do gpt-oss-20b and Llama 3.2 90B share?
3 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Llama 3.2 90B has 9.