Model comparison
gpt-oss-20b vs Llama 2-13B
gpt-oss-20b is the stronger model overall, scoring 32.5 to 29.6 on the Noometry Index.
Last verified . 19 shared benchmarks.
MetaSummary
- They share 19 benchmarks with published results for both. gpt-oss-20b scores higher in 8 categories and Llama 2-13B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where gpt-oss-20b leads 42.2 to 26.5.
- The biggest single-benchmark swing is DTBench: 68% for gpt-oss-20b and 42.2% for Llama 2-13B.
Side by side
| gpt-oss-20b | Llama 2-13B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 32.5 | 29.6 |
| Released | 2025-08-05 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.018 | — |
| Output $ / M tokens | $0.09 | — |
| Results tracked | 34 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
gpt-oss-20b: 37.6 (#192), Llama 2-13B: 30.9 (#291)
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1306 | 1062 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| ALE-Bench | 566.05 | — |
Agentic & Tool Use Not comparable
gpt-oss-20b: 9.3 (#154), Llama 2-13B: —
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| Terminal-Bench | 3.4% | — |
Reasoning gpt-oss-20b leads
gpt-oss-20b: 19.3 (#261), Llama 2-13B: 12.8 (#337)
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| Chess Puzzles | 4% | 0% |
| LMArena Hard Prompts | 1274 | 1051 |
| DTBench | 68% | 42.2% |
| Epoch Capabilities Index | 137.82 | 106.17 |
| Kagi LLM Benchmark | 53.2% | — |
| CritPt | 1.4% | — |
| LMCA | 14.5% | — |
| BIG-Bench Hard | — | 58.2% |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| PIQA | — | 80.8% |
| WinoGrande | — | 72.8% |
Math gpt-oss-20b leads
gpt-oss-20b: 39.4 (#103), Llama 2-13B: 31.1 (#229)
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| LMArena Math | 1317 | 1065 |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| Omni-MATH | 56.5% | — |
| GSM8K | — | 36.9% |
Knowledge gpt-oss-20b leads
gpt-oss-20b: 34.6 (#195), Llama 2-13B: 28.1 (#249)
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1258 | 1030 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
| ARC (AI2) Challenge | — | 60.3% |
| BoolQ | — | 82.4% |
| MMLU | — | 55.6% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |
Multimodal Not comparable
gpt-oss-20b: —, Llama 2-13B: —
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| ScienceQA | — | 55.8% |
Multilingual gpt-oss-20b leads
gpt-oss-20b: 42.2 (#197), Llama 2-13B: 26.5 (#279)
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1268 | 1024 |
| LMArena Chinese | 1314 | 1001 |
| LMArena German | 1255 | 1009 |
| LMArena Japanese | 1244 | 894 |
| LMArena Korean | 1236 | 953 |
| LMArena Russian | 1278 | 1055 |
| LMArena Spanish | 1267 | 1087 |
| LMArena French | — | 1044 |
Instruction Following gpt-oss-20b leads
gpt-oss-20b: 61.8 (#240), Llama 2-13B: 53.3 (#287)
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1236 | 1045 |
| IFEval | 73.2% | — |
Long Context gpt-oss-20b leads
gpt-oss-20b: 37.9 (#209), Llama 2-13B: 32.3 (#269)
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1250 | 1064 |
Writing & Preference gpt-oss-20b leads
gpt-oss-20b: 35.5 (#265), Llama 2-13B: 29.8 (#289)
| Benchmark | gpt-oss-20b | Llama 2-13B |
|---|---|---|
| LMArena Text | 1287 | 1084 |
| LMArena Creative Writing | 1201 | 1047 |
| LMArena Multi-Turn | 1268 | 1050 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
Frequently asked questions
Is gpt-oss-20b better than Llama 2-13B?
gpt-oss-20b is the stronger model overall, scoring 32.5 to 29.6 on the Noometry Index.
Is gpt-oss-20b or Llama 2-13B better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.6 versus 30.9 in the Noometry coding category.
How many benchmarks do gpt-oss-20b and Llama 2-13B share?
19 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Llama 2-13B has 32.