Model comparison
Hy3 vs Llama 3.1-405B
Hy3 is the stronger model overall, scoring 43.6 to 30.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
Tencent
MetaSummary
- They share 15 benchmarks with published results for both. Hy3 scores higher in 8 categories and Llama 3.1-405B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Hy3 leads 62.5 to 39.1.
Side by side
| Hy3 | Llama 3.1-405B | |
|---|---|---|
| Provider | Tencent | Meta |
| Noometry Index | 43.6 | 30.4 |
| Released | 2026-07-06 | 2024-07-23 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.13 | — |
| Output $ / M tokens | $0.53 | — |
| Results tracked | 17 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
Hy3: 46.6 (#60), Llama 3.1-405B: 32.7 (#257)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| LMArena Coding | 1465 | 1291 |
| LMArena WebDev | 1509 | — |
| WeirdML | — | 21.4% |
Agentic & Tool Use Not comparable
Hy3: —, Llama 3.1-405B: 20.9 (#131)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| TheAgentCompany | — | 7.4% |
| Cybench | — | 7.5% |
Reasoning Hy3 leads
Hy3: 25.0 (#134), Llama 3.1-405B: 16.4 (#293)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| LMArena Hard Prompts | 1449 | 1270 |
| SimpleBench | — | 23% |
| Kagi LLM Benchmark | — | 45% |
| NYT Connections (extended) | 41.2% | — |
| DTBench | — | 61.4% |
| BIG-Bench Hard | — | 82.9% |
| Epoch Capabilities Index | — | 128.76 |
| ForecastBench | — | 59.9 |
| HellaSwag | — | 89.2% |
| PIQA | — | 85.9% |
| WinoGrande | — | 89.2% |
Math Hy3 leads
Hy3: 37.4 (#98), Llama 3.1-405B: 17.3 (#283)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| LMArena Math | 1475 | 1281 |
| OTIS Mock AIME 2024-2025 | — | 9.7% |
| Omni-MATH | — | 24.9% |
| MATH Level 5 | — | 49.8% |
Knowledge Hy3 leads
Hy3: 40.2 (#108), Llama 3.1-405B: 30.4 (#221)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| LMArena Expert | 1461 | 1243 |
| GPQA Diamond | — | 50.9% |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 17.6% |
| GPQA (HELM) | — | 52.2% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 84.5% |
| TriviaQA | — | 82.7% |
Multilingual Hy3 leads
Hy3: 53.7 (#60), Llama 3.1-405B: 40.9 (#209)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| LMArena Non-English | 1427 | 1248 |
| LMArena Chinese | 1496 | 1242 |
| LMArena French | 1461 | 1278 |
| LMArena Korean | 1390 | 1184 |
| LMArena Russian | 1435 | 1265 |
| LMArena Spanish | 1448 | 1259 |
| LMArena German | — | 1252 |
| LMArena Japanese | — | 1208 |
Instruction Following Hy3 leads
Hy3: 75.3 (#68), Llama 3.1-405B: 66.1 (#209)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| LMArena Instruction Following | 1425 | 1259 |
| IFEval | — | 81.1% |
Long Context Hy3 leads
Hy3: 44.1 (#70), Llama 3.1-405B: 38.4 (#192)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| LMArena Longer Query | 1443 | 1266 |
Writing & Preference Hy3 leads
Hy3: 62.5 (#75), Llama 3.1-405B: 39.1 (#246)
| Benchmark | Hy3 | Llama 3.1-405B |
|---|---|---|
| LMArena Text | 1442 | 1284 |
| LMArena Creative Writing | 1399 | 1262 |
| LMArena Multi-Turn | 1436 | 1297 |
| EQ-Bench Creative Writing | — | 870 |
| WildBench | — | 78.3% |
Frequently asked questions
Is Hy3 better than Llama 3.1-405B?
Hy3 is the stronger model overall, scoring 43.6 to 30.4 on the Noometry Index.
Is Hy3 or Llama 3.1-405B better for coding?
Hy3 scores higher on coding benchmarks: 46.6 versus 32.7 in the Noometry coding category.
How many benchmarks do Hy3 and Llama 3.1-405B share?
15 benchmarks have published results for both models. Hy3 has 17 scored results on Noometry and Llama 3.1-405B has 42.