Model comparison
GPT-5.6 Sol vs Llama 3.1-405B
GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 30.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
MetaSummary
- They share 25 benchmarks with published results for both. GPT-5.6 Sol scores higher in 9 categories and Llama 3.1-405B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Sol leads 85.6 to 18.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for GPT-5.6 Sol and 9.7% for Llama 3.1-405B.
- Llama 3.1-405B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.6 Sol | Llama 3.1-405B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 65.0 | 30.7 |
| Released | 2026-07-09 | 2024-07-23 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 128K | — |
| Input $ / M tokens | $4 | — |
| Output $ / M tokens | $20 | — |
| Results tracked | 65 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Sol leads
GPT-5.6 Sol: 65.1 (#7), Llama 3.1-405B: 33.1 (#262)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| WeirdML | 89.4% | 21.4% |
| LMArena Coding | 1498 | 1291 |
| DeepSWE | 72.7% | — |
| FrontierCode | 47.5% | — |
| CursorBench | 41.7% | — |
| LMArena WebDev | 1618 | — |
| FrontierSWE | 32.2% | — |
| SciCode | 57.1% | — |
| GSO | 76.5% | — |
| MirrorCode | 20% | — |
| ALE-Bench | 2,177 | — |
Agentic & Tool Use GPT-5.6 Sol leads
GPT-5.6 Sol: 50.3 (#7), Llama 3.1-405B: 21.0 (#140)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| APEX-Agents | 51.4% | — |
| OSWorld 2.0 | 27.3% | — |
| TheAgentCompany | — | 7.4% |
| τ²-bench Banking | 46.9% | — |
| Cybench | — | 7.5% |
| PostTrainBench | 36.2% | — |
| BALROG | 60% | — |
| GBAEval | 52.6% | — |
| GDP.pdf | 30.7% | — |
| LMArena Search | 1257 | — |
| Vending-Bench 2 | 9,619 | — |
Reasoning GPT-5.6 Sol leads
GPT-5.6 Sol: 74.8 (#8), Llama 3.1-405B: 16.8 (#300)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| SimpleBench | 71.7% | 23% |
| Kagi LLM Benchmark | 67% | 45% |
| LMArena Hard Prompts | 1484 | 1269 |
| DTBench | 96% | 61.4% |
| Epoch Capabilities Index | 161.66 | 128.75 |
| ARC-AGI-2 | 92.5% | — |
| NYT Connections (extended) | 93.8% | — |
| ARC-AGI-1 | 97.5% | — |
| CritPt | 32.3% | — |
| Chess Puzzles | 64% | — |
| EnigmaEval | 37.1% | — |
| EBR-Bench | 44.8% | — |
| Mystery Game Puzzles | 58% | — |
| LMCA | 59.2% | — |
| Surface Evolver Bench | 93.1% | — |
| Bench to the Future 3 | 0.14 | — |
| BIG-Bench Hard | — | 82.9% |
| ForecastBench | — | 59.9 |
| HellaSwag | — | 89.2% |
| PIQA | — | 85.9% |
| WinoGrande | — | 89.2% |
Math GPT-5.6 Sol leads
GPT-5.6 Sol: 85.6 (#9), Llama 3.1-405B: 18.4 (#290)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 9.7% |
| LMArena Math | 1474 | 1281 |
| FrontierMath (Tiers 1-3) | 89.1% | — |
| FrontierMath Tier 4 | 82.9% | — |
| ProofBench | 83% | — |
| Omni-MATH | — | 24.9% |
| MATH Level 5 | — | 49.8% |
| FrontierMath Erdős | 0% | — |
Knowledge GPT-5.6 Sol leads
GPT-5.6 Sol: 64.3 (#18), Llama 3.1-405B: 30.4 (#227)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| GPQA Diamond | 93.5% | 50.9% |
| LMArena Expert | 1516 | 1243 |
| SimpleQA Verified | 69.7% | — |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 17.6% |
| Vectara Hallucination Rate | 12.4% | — |
| GPQA (HELM) | — | 52.2% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 84.5% |
| TriviaQA | — | 82.7% |
Multimodal Not comparable
GPT-5.6 Sol: 48.6 (#9), Llama 3.1-405B: —
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| LMArena Vision | 1281 | — |
| Blueprint-Bench 2 | 33.6% | — |
| Furniture Assembly | 56.7% | — |
| LMArena Document | 1483 | — |
Multilingual GPT-5.6 Sol leads
GPT-5.6 Sol: 55.3 (#32), Llama 3.1-405B: 40.7 (#214)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| LMArena Non-English | 1452 | 1248 |
| LMArena Chinese | 1527 | 1242 |
| LMArena French | 1477 | 1279 |
| LMArena German | 1476 | 1252 |
| LMArena Japanese | 1471 | 1208 |
| LMArena Korean | 1442 | 1184 |
| LMArena Russian | 1468 | 1265 |
| LMArena Spanish | 1441 | 1260 |
Instruction Following GPT-5.6 Sol leads
GPT-5.6 Sol: 77.7 (#16), Llama 3.1-405B: 65.9 (#214)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| LMArena Instruction Following | 1482 | 1259 |
| IFEval | — | 81.1% |
Long Context GPT-5.6 Sol leads
GPT-5.6 Sol: 45.4 (#42), Llama 3.1-405B: 38.4 (#197)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| LMArena Longer Query | 1480 | 1266 |
Writing & Preference GPT-5.6 Sol leads
GPT-5.6 Sol: 73.3 (#12), Llama 3.1-405B: 38.9 (#251)
| Benchmark | GPT-5.6 Sol | Llama 3.1-405B |
|---|---|---|
| LMArena Text | 1457 | 1284 |
| LMArena Creative Writing | 1448 | 1262 |
| EQ-Bench Creative Writing | 1972 | 870 |
| LMArena Multi-Turn | 1460 | 1297 |
| WildBench | — | 78.3% |
| EQ-Bench 4 | 1250 | — |
Frequently asked questions
Is GPT-5.6 Sol better than Llama 3.1-405B?
GPT-5.6 Sol is the stronger model overall, scoring 65.0 to 30.7 on the Noometry Index.
Is GPT-5.6 Sol or Llama 3.1-405B better for coding?
GPT-5.6 Sol scores higher on coding benchmarks: 65.1 versus 33.1 in the Noometry coding category.
How many benchmarks do GPT-5.6 Sol and Llama 3.1-405B share?
25 benchmarks have published results for both models. GPT-5.6 Sol has 65 scored results on Noometry and Llama 3.1-405B has 42.