Model comparison
DeepSeek-R1-Distill-Llama-70B vs Hy3
Hy3 is the stronger model overall, scoring 44.2 to 37.8 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeek
TencentSummary
- The widest gap is in writing & preference, where Hy3 leads 62.2 to 49.0.
Side by side
| DeepSeek-R1-Distill-Llama-70B | Hy3 | |
|---|---|---|
| Provider | DeepSeek | Tencent |
| Noometry Index | 37.8 | 44.2 |
| Released | 2025-01-20 | 2026-07-06 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.13 |
| Output $ / M tokens | — | $0.53 |
| Results tracked | 13 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
DeepSeek-R1-Distill-Llama-70B: 36.8 (#202), Hy3: 46.8 (#63)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Hy3 |
|---|---|---|
| LMArena WebDev | — | 1508 |
| BigCodeBench Instruct | 35.3% | — |
| LiveBench Coding | 51.6% | — |
| LMArena Coding | — | 1464 |
| BigCodeBench Complete | 49.9% | — |
Reasoning Hy3 leads
DeepSeek-R1-Distill-Llama-70B: 24.9 (#156), Hy3: 26.1 (#136)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Hy3 |
|---|---|---|
| Kagi LLM Benchmark | 52.3% | — |
| NYT Connections (extended) | — | 41.2% |
| LiveBench Reasoning | 67.6% | — |
| LMArena Hard Prompts | — | 1447 |
| LiveBench Data Analysis | 55.9% | — |
| LiveBench | 54.5% | — |
Math Hy3 leads
DeepSeek-R1-Distill-Llama-70B: 36.0 (#176), Hy3: 40.1 (#93)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Hy3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 51.4% | — |
| LiveBench Math | 58.1% | — |
| LMArena Math | — | 1475 |
| MATH Level 5 | 89.9% | — |
Knowledge Hy3 leads
DeepSeek-R1-Distill-Llama-70B: 30.7 (#225), Hy3: 40.8 (#114)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Hy3 |
|---|---|---|
| GPQA Diamond | 55.7% | — |
| LMArena Expert | — | 1460 |
Multilingual Not comparable
DeepSeek-R1-Distill-Llama-70B: —, Hy3: 53.5 (#65)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Hy3 |
|---|---|---|
| LMArena Non-English | — | 1426 |
| LMArena Chinese | — | 1493 |
| LMArena French | — | 1461 |
| LMArena German | — | 1439 |
| LMArena Japanese | — | 1392 |
| LMArena Korean | — | 1395 |
| LMArena Russian | — | 1432 |
| LMArena Spanish | — | 1456 |
Instruction Following Hy3 leads
DeepSeek-R1-Distill-Llama-70B: 68.2 (#190), Hy3: 75.1 (#70)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Hy3 |
|---|---|---|
| LiveBench Instruction Following | 69.9% | — |
| LMArena Instruction Following | — | 1426 |
Long Context Not comparable
DeepSeek-R1-Distill-Llama-70B: —, Hy3: 44.1 (#75)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Hy3 |
|---|---|---|
| LMArena Longer Query | — | 1442 |
Writing & Preference Hy3 leads
DeepSeek-R1-Distill-Llama-70B: 49.0 (#194), Hy3: 62.2 (#81)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Hy3 |
|---|---|---|
| LMArena Text | — | 1439 |
| LMArena Creative Writing | — | 1402 |
| LMArena Multi-Turn | — | 1436 |
| LiveBench Language | 23.8% | — |
Frequently asked questions
Is DeepSeek-R1-Distill-Llama-70B better than Hy3?
Hy3 is the stronger model overall, scoring 44.2 to 37.8 on the Noometry Index.
Is DeepSeek-R1-Distill-Llama-70B or Hy3 better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 36.8 in the Noometry coding category.
How many benchmarks do DeepSeek-R1-Distill-Llama-70B and Hy3 share?
0 benchmarks have published results for both models. DeepSeek-R1-Distill-Llama-70B has 13 scored results on Noometry and Hy3 has 19.