Model comparison
gpt-oss-120b vs Llama 3-70B
gpt-oss-120b is the stronger model overall, scoring 36.3 to 28.8 on the Noometry Index.
Last verified . 22 shared benchmarks.
MetaSummary
- They share 22 benchmarks with published results for both. gpt-oss-120b scores higher in 6 categories and Llama 3-70B in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 52.5 to 12.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for gpt-oss-120b and 4.3% for Llama 3-70B.
Side by side
| gpt-oss-120b | Llama 3-70B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 36.3 | 28.8 |
| Released | 2025-08-05 | 2024-04-18 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 41K | — |
| Input $ / M tokens | $0.037 | — |
| Output $ / M tokens | $0.17 | — |
| Results tracked | 48 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-70B leads
gpt-oss-120b: 33.5 (#256), Llama 3-70B: 35.8 (#218)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1380 | 1206 |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| SciCode | 36% | — |
| WeirdML | 48.2% | — |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use Llama 3-70B leads
gpt-oss-120b: 12.2 (#153), Llama 3-70B: 21.1 (#139)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| Cybench | — | 5% |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning gpt-oss-120b leads
gpt-oss-120b: 20.0 (#245), Llama 3-70B: 18.0 (#288)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| Kagi LLM Benchmark | 58.6% | 35.1% |
| LMArena Hard Prompts | 1364 | 1195 |
| DTBench | 76.3% | 54.2% |
| Epoch Capabilities Index | 139.93 | 122.93 |
| SimpleBench | 22.1% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| Mystery Game Puzzles | 2% | — |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |
Math gpt-oss-120b leads
gpt-oss-120b: 52.5 (#50), Llama 3-70B: 12.8 (#305)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 4.3% |
| LMArena Math | 1389 | 1218 |
| Omni-MATH | 68.8% | — |
| MATH Level 5 | — | 22.6% |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.4 (#96), Llama 3-70B: 20.8 (#277)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 75.8% | 40.6% |
| LMArena Expert | 1356 | 1149 |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |
| MMLU | — | 79.3% |
Multilingual gpt-oss-120b leads
gpt-oss-120b: 48.0 (#147), Llama 3-70B: 33.6 (#251)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1351 | 1142 |
| LMArena Chinese | 1385 | 1114 |
| LMArena French | 1369 | 1232 |
| LMArena German | 1353 | 1169 |
| LMArena Japanese | 1331 | 1017 |
| LMArena Korean | 1282 | 1017 |
| LMArena Russian | 1343 | 1159 |
| LMArena Spanish | 1389 | 1241 |
Instruction Following gpt-oss-120b leads
gpt-oss-120b: 69.3 (#173), Llama 3-70B: 62.5 (#238)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1318 | 1194 |
| IFEval | 83.6% | — |
Long Context Llama 3-70B leads
gpt-oss-120b: 31.4 (#278), Llama 3-70B: 35.6 (#240)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1319 | 1174 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference gpt-oss-120b leads
gpt-oss-120b: 46.5 (#217), Llama 3-70B: 42.8 (#231)
| Benchmark | gpt-oss-120b | Llama 3-70B |
|---|---|---|
| LMArena Text | 1365 | 1221 |
| LMArena Creative Writing | 1275 | 1210 |
| LMArena Multi-Turn | 1340 | 1223 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
Frequently asked questions
Is gpt-oss-120b better than Llama 3-70B?
gpt-oss-120b is the stronger model overall, scoring 36.3 to 28.8 on the Noometry Index.
Is gpt-oss-120b or Llama 3-70B better for coding?
Llama 3-70B scores higher on coding benchmarks: 35.8 versus 33.5 in the Noometry coding category.
How many benchmarks do gpt-oss-120b and Llama 3-70B share?
22 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Llama 3-70B has 31.