Model comparison
Hy3 vs Llama 3.2 90B
Hy3 is the stronger model overall, scoring 43.6 to 27.1 on the Noometry Index.
Last verified . 0 shared benchmarks.
Tencent
MetaSummary
- The widest gap is in math, where Hy3 leads 37.4 to 10.4.
Side by side
| Hy3 | Llama 3.2 90B | |
|---|---|---|
| Provider | Tencent | Meta |
| Noometry Index | 43.6 | 27.1 |
| Released | 2026-07-06 | 2024-09-24 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.13 | — |
| Output $ / M tokens | $0.53 | — |
| Results tracked | 17 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Hy3: 46.6 (#60), Llama 3.2 90B: —
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| LMArena WebDev | 1509 | — |
| LMArena Coding | 1465 | — |
Agentic & Tool Use Not comparable
Hy3: —, Llama 3.2 90B: 29.7 (#73)
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| BALROG | — | 27.3% |
Reasoning Hy3 leads
Hy3: 25.0 (#134), Llama 3.2 90B: 20.9 (#207)
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| NYT Connections (extended) | 41.2% | — |
| EnigmaEval | — | 0.4% |
| LMArena Hard Prompts | 1449 | — |
| Epoch Capabilities Index | — | 125.5 |
Math Hy3 leads
Hy3: 37.4 (#98), Llama 3.2 90B: 10.4 (#302)
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 2.6% |
| LMArena Math | 1475 | — |
| MATH Level 5 | — | 39.4% |
Knowledge Hy3 leads
Hy3: 40.2 (#108), Llama 3.2 90B: 21.7 (#269)
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| GPQA Diamond | — | 41% |
| LMArena Expert | 1461 | — |
| MMLU | — | 80.3% |
Multimodal Not comparable
Hy3: —, Llama 3.2 90B: 24.9 (#119)
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| LMArena Vision | — | 1000 |
| GeoBench | — | 52% |
Multilingual Not comparable
Hy3: 53.7 (#60), Llama 3.2 90B: —
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| LMArena Non-English | 1427 | — |
| LMArena Chinese | 1496 | — |
| LMArena French | 1461 | — |
| LMArena Korean | 1390 | — |
| LMArena Russian | 1435 | — |
| LMArena Spanish | 1448 | — |
Instruction Following Not comparable
Hy3: 75.3 (#68), Llama 3.2 90B: —
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| LMArena Instruction Following | 1425 | — |
Long Context Not comparable
Hy3: 44.1 (#70), Llama 3.2 90B: —
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| LMArena Longer Query | 1443 | — |
Writing & Preference Not comparable
Hy3: 62.5 (#75), Llama 3.2 90B: —
| Benchmark | Hy3 | Llama 3.2 90B |
|---|---|---|
| LMArena Text | 1442 | — |
| LMArena Creative Writing | 1399 | — |
| LMArena Multi-Turn | 1436 | — |
Frequently asked questions
Is Hy3 better than Llama 3.2 90B?
Hy3 is the stronger model overall, scoring 43.6 to 27.1 on the Noometry Index.
How many benchmarks do Hy3 and Llama 3.2 90B share?
0 benchmarks have published results for both models. Hy3 has 17 scored results on Noometry and Llama 3.2 90B has 9.