Model comparison
Hunyuan Large 2025 02 10 vs Llama 3-70B
Hunyuan Large 2025 02 10 is the stronger model overall, scoring 38.0 to 28.5 on the Noometry Index.
Last verified . 12 shared benchmarks.
Tencent
MetaSummary
- They share 12 benchmarks with published results for both. Hunyuan Large 2025 02 10 scores higher in 8 categories and Llama 3-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Hunyuan Large 2025 02 10 leads 33.1 to 12.0.
- Llama 3-70B has downloadable open weights; the other is API-only.
Side by side
| Hunyuan Large 2025 02 10 | Llama 3-70B | |
|---|---|---|
| Provider | Tencent | Meta |
| Noometry Index | 38.0 | 28.5 |
| Released | — | 2024-04-18 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 12 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hunyuan Large 2025 02 10 leads
Hunyuan Large 2025 02 10: 37.6 (#177), Llama 3-70B: 35.3 (#215)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1307 | 1206 |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use Not comparable
Hunyuan Large 2025 02 10: —, Llama 3-70B: 20.9 (#132)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| Cybench | — | 5% |
Reasoning Hunyuan Large 2025 02 10 leads
Hunyuan Large 2025 02 10: 24.5 (#147), Llama 3-70B: 17.4 (#284)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 1286 | 1195 |
| Kagi LLM Benchmark | — | 35.1% |
| DTBench | — | 54.2% |
| Epoch Capabilities Index | — | 122.91 |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |
Math Hunyuan Large 2025 02 10 leads
Hunyuan Large 2025 02 10: 33.1 (#178), Llama 3-70B: 12.0 (#299)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| LMArena Math | 1281 | 1218 |
| OTIS Mock AIME 2024-2025 | — | 4.3% |
| MATH Level 5 | — | 22.6% |
Knowledge Hunyuan Large 2025 02 10 leads
Hunyuan Large 2025 02 10: 34.5 (#188), Llama 3-70B: 20.8 (#272)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| LMArena Expert | 1276 | 1149 |
| GPQA Diamond | — | 40.6% |
| MMLU | — | 79.3% |
Multilingual Hunyuan Large 2025 02 10 leads
Hunyuan Large 2025 02 10: 42.1 (#195), Llama 3-70B: 33.8 (#246)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1265 | 1142 |
| LMArena Chinese | 1346 | 1114 |
| LMArena Russian | 1266 | 1159 |
| LMArena French | — | 1231 |
| LMArena German | — | 1169 |
| LMArena Japanese | — | 1017 |
| LMArena Korean | — | 1017 |
| LMArena Spanish | — | 1241 |
Instruction Following Hunyuan Large 2025 02 10 leads
Hunyuan Large 2025 02 10: 67.6 (#192), Llama 3-70B: 62.8 (#233)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1277 | 1194 |
Long Context Hunyuan Large 2025 02 10 leads
Hunyuan Large 2025 02 10: 40.8 (#144), Llama 3-70B: 35.6 (#235)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1342 | 1174 |
Writing & Preference Hunyuan Large 2025 02 10 leads
Hunyuan Large 2025 02 10: 48.9 (#191), Llama 3-70B: 43.0 (#225)
| Benchmark | Hunyuan Large 2025 02 10 | Llama 3-70B |
|---|---|---|
| LMArena Text | 1288 | 1221 |
| LMArena Creative Writing | 1264 | 1210 |
| LMArena Multi-Turn | 1284 | 1223 |
Frequently asked questions
Is Hunyuan Large 2025 02 10 better than Llama 3-70B?
Hunyuan Large 2025 02 10 is the stronger model overall, scoring 38.0 to 28.5 on the Noometry Index.
Is Hunyuan Large 2025 02 10 or Llama 3-70B better for coding?
Hunyuan Large 2025 02 10 scores higher on coding benchmarks: 37.6 versus 35.3 in the Noometry coding category.
How many benchmarks do Hunyuan Large 2025 02 10 and Llama 3-70B share?
12 benchmarks have published results for both models. Hunyuan Large 2025 02 10 has 12 scored results on Noometry and Llama 3-70B has 31.