Model comparison
GPT-5.6 Luna vs Llama 2-70B
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 24.4 on the Noometry Index.
Last verified . 21 shared benchmarks.
MetaSummary
- They share 21 benchmarks with published results for both. GPT-5.6 Luna scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 8.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.3% for GPT-5.6 Luna and 0% for Llama 2-70B.
- Llama 2-70B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.6 Luna | Llama 2-70B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 54.6 | 24.4 |
| Released | 2026-07-09 | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.20 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 52 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
GPT-5.6 Luna: 54.5 (#28), Llama 2-70B: 31.4 (#286)
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1466 | 1079 |
| DeepSWE | 67.2% | — |
| FrontierCode | 39.8% | — |
| CursorBench | 35.9% | — |
| LMArena WebDev | 1519 | — |
| SciCode | 53.6% | — |
| WeirdML | 60.9% | — |
| ALE-Bench | 1,667 | — |
Agentic & Tool Use Not comparable
GPT-5.6 Luna: 34.4 (#45), Llama 2-70B: —
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| APEX-Agents | 43% | — |
| BALROG | 45.6% | — |
| GDP.pdf | 22.7% | — |
| Vending-Bench 2 | 4,095 | — |
Reasoning GPT-5.6 Luna leads
GPT-5.6 Luna: 47.6 (#43), Llama 2-70B: 14.4 (#325)
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1451 | 1073 |
| DTBench | 89.1% | 41.6% |
| Epoch Capabilities Index | 156.39 | 113.79 |
| ARC-AGI-2 | 59.5% | — |
| SimpleBench | 46.8% | — |
| Kagi LLM Benchmark | 49.1% | — |
| NYT Connections (extended) | 69.4% | — |
| ARC-AGI-1 | 88% | — |
| CritPt | 20.6% | — |
| Chess Puzzles | 40% | — |
| Mystery Game Puzzles | 21% | — |
| LMCA | 48.5% | — |
| Surface Evolver Bench | 61.9% | — |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math GPT-5.6 Luna leads
GPT-5.6 Luna: 77.7 (#14), Llama 2-70B: 8.1 (#326)
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 0% |
| LMArena Math | 1458 | 1091 |
| FrontierMath (Tiers 1-3) | 82.1% | — |
| FrontierMath Tier 4 | 61% | — |
| ProofBench | 60% | — |
| MATH Level 5 | — | 3.3% |
| GSM8K | — | 69.6% |
Knowledge GPT-5.6 Luna leads
GPT-5.6 Luna: 58.5 (#34), Llama 2-70B: 7.4 (#310)
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| GPQA Diamond | 91.6% | 26.3% |
| LMArena Expert | 1478 | 1039 |
| SimpleQA Verified | 41% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multimodal Not comparable
GPT-5.6 Luna: 42.7 (#28), Llama 2-70B: —
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| LMArena Vision | 1258 | — |
| Blueprint-Bench 2 | 22.6% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1457 | — |
Multilingual GPT-5.6 Luna leads
GPT-5.6 Luna: 52.8 (#78), Llama 2-70B: 27.7 (#274)
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1417 | 1045 |
| LMArena Chinese | 1470 | 995 |
| LMArena French | 1456 | 1090 |
| LMArena German | 1454 | 1041 |
| LMArena Japanese | 1411 | 927 |
| LMArena Korean | 1415 | 964 |
| LMArena Russian | 1428 | 1083 |
| LMArena Spanish | 1448 | 1143 |
Instruction Following GPT-5.6 Luna leads
GPT-5.6 Luna: 75.6 (#57), Llama 2-70B: 54.9 (#278)
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1437 | 1071 |
Long Context GPT-5.6 Luna leads
GPT-5.6 Luna: 43.9 (#82), Llama 2-70B: 32.3 (#270)
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1436 | 1062 |
Writing & Preference GPT-5.6 Luna leads
GPT-5.6 Luna: 68.0 (#29), Llama 2-70B: 32.3 (#279)
| Benchmark | GPT-5.6 Luna | Llama 2-70B |
|---|---|---|
| LMArena Text | 1431 | 1115 |
| LMArena Creative Writing | 1396 | 1075 |
| LMArena Multi-Turn | 1434 | 1088 |
| EQ-Bench Creative Writing | 1829 | — |
| EQ-Bench 4 | 1156 | — |
Frequently asked questions
Is GPT-5.6 Luna better than Llama 2-70B?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 24.4 on the Noometry Index.
Is GPT-5.6 Luna or Llama 2-70B better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 31.4 in the Noometry coding category.
How many benchmarks do GPT-5.6 Luna and Llama 2-70B share?
21 benchmarks have published results for both models. GPT-5.6 Luna has 52 scored results on Noometry and Llama 2-70B has 35.