Model comparison
Llama 3.1-405B vs o1-mini
o1-mini is the stronger model overall, scoring 34.0 to 30.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
MetaSummary
- They share 25 benchmarks with published results for both. Llama 3.1-405B scores higher in 1 category and o1-mini in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where o1-mini leads 35.4 to 18.4.
- The biggest single-benchmark swing is MATH Level 5: 49.8% for Llama 3.1-405B and 89.2% for o1-mini.
- Llama 3.1-405B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.1-405B | o1-mini | |
|---|---|---|
| Provider | Meta | OpenAI |
| Noometry Index | 30.7 | 34.0 |
| Released | 2024-07-23 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 42 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1-mini leads
Llama 3.1-405B: 33.1 (#262), o1-mini: 35.5 (#224)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| WeirdML | 21.4% | 36.3% |
| LMArena Coding | 1291 | 1362 |
| Aider Polyglot | — | 32.9% |
| LiveBench Coding | — | 48% |
| HumanEval+ | — | 89% |
| MBPP+ | — | 78.8% |
Agentic & Tool Use o1-mini leads
Llama 3.1-405B: 21.0 (#140), o1-mini: 24.6 (#118)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| Cybench | 7.5% | 10% |
| TheAgentCompany | 7.4% | — |
Reasoning Llama 3.1-405B leads
Llama 3.1-405B: 16.8 (#300), o1-mini: 8.8 (#346)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| SimpleBench | 23% | 18.1% |
| LMArena Hard Prompts | 1269 | 1333 |
| Epoch Capabilities Index | 128.75 | 135.82 |
| ARC-AGI-2 | — | 0.8% |
| Kagi LLM Benchmark | 45% | — |
| ARC-AGI-1 | — | 14% |
| LiveBench Reasoning | — | 72.3% |
| DTBench | 61.4% | — |
| LiveBench Data Analysis | — | 57.9% |
| BIG-Bench Hard | 82.9% | — |
| ForecastBench | 59.9 | — |
| HellaSwag | 89.2% | — |
| LiveBench | — | 57.8% |
| PIQA | 85.9% | — |
| WinoGrande | 89.2% | — |
Math o1-mini leads
Llama 3.1-405B: 18.4 (#290), o1-mini: 35.4 (#186)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 9.7% | 46.9% |
| LMArena Math | 1281 | 1358 |
| MATH Level 5 | 49.8% | 89.2% |
| Omni-MATH | 24.9% | — |
| LiveBench Math | — | 62% |
| FrontierMath (Feb 2025 set) | — | 1.7% |
Knowledge o1-mini leads
Llama 3.1-405B: 30.4 (#227), o1-mini: 34.9 (#192)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| GPQA Diamond | 50.9% | 62.4% |
| Confabulations | 17.6% | 18.6% |
| LMArena Expert | 1243 | 1316 |
| MMLU-Pro | 72.3% | — |
| GPQA (HELM) | 52.2% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 84.5% | — |
| TriviaQA | 82.7% | — |
Multilingual o1-mini leads
Llama 3.1-405B: 40.7 (#214), o1-mini: 43.6 (#182)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| LMArena Non-English | 1248 | 1289 |
| LMArena Chinese | 1242 | 1314 |
| LMArena French | 1279 | 1293 |
| LMArena German | 1252 | 1278 |
| LMArena Japanese | 1208 | 1245 |
| LMArena Korean | 1184 | 1223 |
| LMArena Russian | 1265 | 1283 |
| LMArena Spanish | 1260 | 1303 |
Instruction Following Too close to call
Llama 3.1-405B: 65.9 (#214), o1-mini: 66.7 (#206)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| LMArena Instruction Following | 1259 | 1304 |
| LiveBench Instruction Following | — | 65.4% |
| IFEval | 81.1% | — |
Long Context o1-mini leads
Llama 3.1-405B: 38.4 (#197), o1-mini: 40.1 (#161)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| LMArena Longer Query | 1266 | 1320 |
Writing & Preference o1-mini leads
Llama 3.1-405B: 38.9 (#251), o1-mini: 48.4 (#202)
| Benchmark | Llama 3.1-405B | o1-mini |
|---|---|---|
| LMArena Text | 1284 | 1317 |
| LMArena Creative Writing | 1262 | 1244 |
| LMArena Multi-Turn | 1297 | 1314 |
| Short-Story Creative Writing | — | 64.9% |
| EQ-Bench Creative Writing | 870 | — |
| WildBench | 78.3% | — |
| LiveBench Language | — | 40.9% |
Frequently asked questions
Is Llama 3.1-405B better than o1-mini?
o1-mini is the stronger model overall, scoring 34.0 to 30.7 on the Noometry Index.
Is Llama 3.1-405B or o1-mini better for coding?
o1-mini scores higher on coding benchmarks: 35.5 versus 33.1 in the Noometry coding category.
How many benchmarks do Llama 3.1-405B and o1-mini share?
25 benchmarks have published results for both models. Llama 3.1-405B has 42 scored results on Noometry and o1-mini has 39.