Model comparison
GPT-5.1 vs Llama 3-70B
GPT-5.1 is the stronger model overall, scoring 49.0 to 28.8 on the Noometry Index.
Last verified . 22 shared benchmarks.
MetaSummary
- They share 22 benchmarks with published results for both. GPT-5.1 scores higher in 9 categories and Llama 3-70B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.1 leads 52.2 to 12.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.6% for GPT-5.1 and 4.3% for Llama 3-70B.
- Llama 3-70B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.1 | Llama 3-70B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 49.0 | 28.8 |
| Released | 2025-11-13 | 2024-04-18 |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $1.25 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 63 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
GPT-5.1: 46.4 (#66), Llama 3-70B: 35.8 (#218)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1454 | 1206 |
| SWE-bench Verified | 68% | — |
| SWE-bench Verified (bash only) | 66% | — |
| LMArena WebDev | 1395 | — |
| SciCode | 43.3% | — |
| GSO | 13.7% | — |
| WeirdML | 60.8% | — |
| BigCodeBench Instruct | — | 43.6% |
| LiveBench Coding | 72.5% | — |
| BigCodeBench Complete | — | 54.5% |
| ALE-Bench | 1,192 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use GPT-5.1 leads
GPT-5.1: 32.7 (#60), Llama 3-70B: 21.1 (#139)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| Terminal-Bench | 47.6% | — |
| Cybench | — | 5% |
| DeepResearch Bench | 42.8% | — |
| LMArena Search | 1199 | — |
| Vending-Bench 2 | 1,473 | — |
Reasoning GPT-5.1 leads
GPT-5.1: 39.8 (#58), Llama 3-70B: 18.0 (#288)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 1457 | 1195 |
| DTBench | 90.1% | 54.2% |
| Epoch Capabilities Index | 149.64 | 122.93 |
| ForecastBench | 58.1 | 57.1 |
| ARC-AGI-2 | 17.6% | — |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | — | 35.1% |
| ARC-AGI-1 | 72.8% | — |
| CritPt | 4.9% | — |
| Chess Puzzles | 32% | — |
| EnigmaEval | 11.2% | — |
| LiveBench Reasoning | 95.8% | — |
| Mystery Game Puzzles | 19% | — |
| LiveBench Data Analysis | 72.1% | — |
| LMCA | 43.9% | — |
| LiveBench | 78.8% | — |
| WinoGrande | — | 83.5% |
Math GPT-5.1 leads
GPT-5.1: 52.2 (#51), Llama 3-70B: 12.8 (#305)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.6% | 4.3% |
| LMArena Math | 1447 | 1218 |
| Omni-MATH | 46.4% | — |
| LiveBench Math | 94.5% | — |
| MATH Level 5 | — | 22.6% |
| FrontierMath (Feb 2025 set) | 31% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GPT-5.1 leads
GPT-5.1: 50.6 (#71), Llama 3-70B: 20.8 (#277)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 87.6% | 40.6% |
| LMArena Expert | 1470 | 1149 |
| Humanity's Last Exam | 23.7% | — |
| SimpleQA Verified | 48% | — |
| MMLU-Pro | 57.9% | — |
| Vectara Hallucination Rate | 10.9% | — |
| GPQA (HELM) | 44.2% | — |
| MMLU | — | 79.3% |
Multimodal Not comparable
GPT-5.1: 44.8 (#19), Llama 3-70B: —
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| LMArena Vision | 1250 | — |
| VPCT | 58.7% | — |
| LMArena Document | 1403 | — |
Multilingual GPT-5.1 leads
GPT-5.1: 53.8 (#56), Llama 3-70B: 33.6 (#251)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1431 | 1142 |
| LMArena Chinese | 1495 | 1114 |
| LMArena French | 1450 | 1232 |
| LMArena German | 1438 | 1169 |
| LMArena Japanese | 1453 | 1017 |
| LMArena Korean | 1401 | 1017 |
| LMArena Russian | 1435 | 1159 |
| LMArena Spanish | 1433 | 1241 |
Instruction Following GPT-5.1 leads
GPT-5.1: 83.9 (#1), Llama 3-70B: 62.5 (#238)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1443 | 1194 |
| LiveBench Instruction Following | 93.3% | — |
| IFEval | 93.5% | — |
Long Context GPT-5.1 leads
GPT-5.1: 47.6 (#14), Llama 3-70B: 35.6 (#240)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1447 | 1174 |
| CL-bench | 23.7% | — |
| CL-bench Life | 17.3% | — |
Writing & Preference GPT-5.1 leads
GPT-5.1: 64.5 (#55), Llama 3-70B: 42.8 (#231)
| Benchmark | GPT-5.1 | Llama 3-70B |
|---|---|---|
| LMArena Text | 1443 | 1221 |
| LMArena Creative Writing | 1427 | 1210 |
| LMArena Multi-Turn | 1450 | 1223 |
| WildBench | 86.3% | — |
| LiveBench Language | 80.2% | — |
Frequently asked questions
Is GPT-5.1 better than Llama 3-70B?
GPT-5.1 is the stronger model overall, scoring 49.0 to 28.8 on the Noometry Index.
Is GPT-5.1 or Llama 3-70B better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.4 versus 35.8 in the Noometry coding category.
How many benchmarks do GPT-5.1 and Llama 3-70B share?
22 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and Llama 3-70B has 31.