Model comparison
GPT-5.2 vs Llama 3-70B
GPT-5.2 is the stronger model overall, scoring 54.1 to 28.8 on the Noometry Index.
Last verified . 23 shared benchmarks.
MetaSummary
- They share 23 benchmarks with published results for both. GPT-5.2 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-5.2 leads 60.0 to 12.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 96.1% for GPT-5.2 and 4.3% for Llama 3-70B.
- Llama 3-70B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.2 | Llama 3-70B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 54.1 | 28.8 |
| Released | 2025-12-11 | 2024-04-18 |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $1.75 | — |
| Output $ / M tokens | $14 | — |
| Results tracked | 67 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 leads
GPT-5.2: 51.6 (#37), Llama 3-70B: 35.8 (#218)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1447 | 1206 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| LMArena WebDev | 1416 | — |
| SWE-bench Multilingual | 66.7% | — |
| GSO | 27.4% | — |
| WeirdML | 72.2% | — |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| ALE-Bench | 1,294 | — |
| AlgoTune | 2.05 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use GPT-5.2 leads
GPT-5.2: 40.2 (#24), Llama 3-70B: 21.1 (#139)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| Terminal-Bench | 64.9% | — |
| Berkeley Function Calling Leaderboard | 55.9% | — |
| GDPval | 49.7% | — |
| Remote Labor Index | 2.5% | — |
| τ²-bench Airline | 83% | — |
| τ²-bench Banking | 32.2% | — |
| τ²-bench Retail | 81.6% | — |
| τ²-bench Telecom | 89.7% | — |
| Cybench | — | 5% |
| DeepResearch Bench | 41.1% | — |
| LMArena Search | 1207 | — |
| METR Time Horizons | 75.3% | — |
| Vending-Bench 2 | 3,591 | — |
Reasoning GPT-5.2 leads
GPT-5.2: 50.2 (#35), Llama 3-70B: 18.0 (#288)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| Kagi LLM Benchmark | 73.3% | 35.1% |
| LMArena Hard Prompts | 1445 | 1195 |
| DTBench | 90.9% | 54.2% |
| Epoch Capabilities Index | 153.45 | 122.93 |
| ForecastBench | 60.1 | 57.1 |
| ARC-AGI-2 | 52.9% | — |
| SimpleBench | 45.8% | — |
| NYT Connections (extended) | 83.6% | — |
| ARC-AGI-1 | 86.2% | — |
| Chess Puzzles | 49% | — |
| EnigmaEval | 10.4% | — |
| EBR-Bench | 23% | — |
| Mystery Game Puzzles | 23% | — |
| LMCA | 43.9% | — |
| WinoGrande | — | 83.5% |
Math GPT-5.2 leads
GPT-5.2: 60.0 (#38), Llama 3-70B: 12.8 (#305)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 96.1% | 4.3% |
| LMArena Math | 1440 | 1218 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 31.7% | — |
| MathArena Final-Answer Competitions | 72% | — |
| ProofBench | 15% | — |
| MATH Level 5 | — | 22.6% |
| FrontierMath (Feb 2025 set) | 40.7% | — |
| FrontierMath Tier 4 (v1) | 18.8% | — |
Knowledge GPT-5.2 leads
GPT-5.2: 59.3 (#32), Llama 3-70B: 20.8 (#277)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 91.4% | 40.6% |
| LMArena Expert | 1445 | 1149 |
| Humanity's Last Exam | 27.8% | — |
| SimpleQA Verified | 37.1% | — |
| Vectara Hallucination Rate | 8.4% | — |
| MMLU | — | 79.3% |
Multimodal Not comparable
GPT-5.2: 51.3 (#7), Llama 3-70B: —
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| LMArena Vision | 1268 | — |
| VPCT | 84% | — |
| Furniture Assembly | 38.3% | — |
| LMArena Document | 1405 | — |
Multilingual GPT-5.2 leads
GPT-5.2: 53.4 (#67), Llama 3-70B: 33.6 (#251)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1425 | 1142 |
| LMArena Chinese | 1460 | 1114 |
| LMArena French | 1455 | 1232 |
| LMArena German | 1448 | 1169 |
| LMArena Japanese | 1420 | 1017 |
| LMArena Korean | 1392 | 1017 |
| LMArena Russian | 1440 | 1159 |
| LMArena Spanish | 1433 | 1241 |
Instruction Following GPT-5.2 leads
GPT-5.2: 74.7 (#89), Llama 3-70B: 62.5 (#238)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1417 | 1194 |
Long Context GPT-5.2 leads
GPT-5.2: 44.0 (#78), Llama 3-70B: 35.6 (#240)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1428 | 1174 |
| CL-bench | 18.2% | — |
Writing & Preference GPT-5.2 leads
GPT-5.2: 66.8 (#32), Llama 3-70B: 42.8 (#231)
| Benchmark | GPT-5.2 | Llama 3-70B |
|---|---|---|
| LMArena Text | 1439 | 1221 |
| LMArena Creative Writing | 1401 | 1210 |
| LMArena Multi-Turn | 1458 | 1223 |
| EQ-Bench Creative Writing | 1703 | — |
Frequently asked questions
Is GPT-5.2 better than Llama 3-70B?
GPT-5.2 is the stronger model overall, scoring 54.1 to 28.8 on the Noometry Index.
Is GPT-5.2 or Llama 3-70B better for coding?
GPT-5.2 scores higher on coding benchmarks: 51.6 versus 35.8 in the Noometry coding category.
How many benchmarks do GPT-5.2 and Llama 3-70B share?
23 benchmarks have published results for both models. GPT-5.2 has 67 scored results on Noometry and Llama 3-70B has 31.