Model comparison
DeepSeek LLM 67B vs Hy3
Hy3 is the stronger model overall, scoring 44.2 to 24.9 on the Noometry Index.
Last verified . 10 shared benchmarks.
DeepSeek
TencentSummary
- They share 10 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Hy3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Hy3 leads 40.8 to 7.0.
Side by side
| DeepSeek LLM 67B | Hy3 | |
|---|---|---|
| Provider | DeepSeek | Tencent |
| Noometry Index | 24.9 | 44.2 |
| Released | 2023-11-29 | 2026-07-06 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.0825 |
| Output $ / M tokens | — | $0.33 |
| Results tracked | 15 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
DeepSeek LLM 67B: 31.9 (#278), Hy3: 46.8 (#63)
| Benchmark | DeepSeek LLM 67B | Hy3 |
|---|---|---|
| LMArena Coding | 1096 | 1464 |
| LMArena WebDev | — | 1508 |
Reasoning Hy3 leads
DeepSeek LLM 67B: 16.5 (#304), Hy3: 26.1 (#136)
| Benchmark | DeepSeek LLM 67B | Hy3 |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1447 |
| NYT Connections (extended) | — | 41.2% |
| Chess Puzzles | 0% | — |
| Epoch Capabilities Index | 110.5 | — |
Math Hy3 leads
DeepSeek LLM 67B: 8.7 (#324), Hy3: 40.1 (#93)
| Benchmark | DeepSeek LLM 67B | Hy3 |
|---|---|---|
| LMArena Math | 1108 | 1475 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| MATH Level 5 | 6.4% | — |
Knowledge Hy3 leads
DeepSeek LLM 67B: 7.0 (#313), Hy3: 40.8 (#114)
| Benchmark | DeepSeek LLM 67B | Hy3 |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| LMArena Expert | — | 1460 |
Multilingual Hy3 leads
DeepSeek LLM 67B: 29.4 (#267), Hy3: 53.5 (#65)
| Benchmark | DeepSeek LLM 67B | Hy3 |
|---|---|---|
| LMArena Non-English | 1073 | 1426 |
| LMArena Chinese | 1132 | 1493 |
| LMArena French | — | 1461 |
| LMArena German | — | 1439 |
| LMArena Japanese | — | 1392 |
| LMArena Korean | — | 1395 |
| LMArena Russian | — | 1432 |
| LMArena Spanish | — | 1456 |
Instruction Following Hy3 leads
DeepSeek LLM 67B: 55.4 (#277), Hy3: 75.1 (#70)
| Benchmark | DeepSeek LLM 67B | Hy3 |
|---|---|---|
| LMArena Instruction Following | 1079 | 1426 |
Long Context Hy3 leads
DeepSeek LLM 67B: 33.1 (#265), Hy3: 44.1 (#75)
| Benchmark | DeepSeek LLM 67B | Hy3 |
|---|---|---|
| LMArena Longer Query | 1092 | 1442 |
Writing & Preference Hy3 leads
DeepSeek LLM 67B: 31.6 (#282), Hy3: 62.2 (#81)
| Benchmark | DeepSeek LLM 67B | Hy3 |
|---|---|---|
| LMArena Text | 1105 | 1439 |
| LMArena Creative Writing | 1067 | 1402 |
| LMArena Multi-Turn | 1082 | 1436 |
Frequently asked questions
Is DeepSeek LLM 67B better than Hy3?
Hy3 is the stronger model overall, scoring 44.2 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Hy3 better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Hy3 share?
10 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Hy3 has 19.