Model comparison
DeepSeek-V2.5 (Sep 2024) vs Hy4 preview
Hy4 preview is the stronger model overall, scoring 44.6 to 37.0 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeek
TencentSummary
- The widest gap is in math, where Hy4 preview leads 54.5 to 33.3.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Hy4 preview | |
|---|---|---|
| Provider | DeepSeek | Tencent |
| Noometry Index | 37.0 | 44.6 |
| Released | 2024-09-06 | 2026-08-28 |
| Weights | Open | Open |
| Context window | — | 1.05M |
| Max output | — | 64K |
| Input $ / M tokens | — | $0.83 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 22 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy4 preview leads
DeepSeek-V2.5 (Sep 2024): 31.3 (#274), Hy4 preview: 51.4 (#36)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Hy4 preview |
|---|---|---|
| Aider Polyglot | 17.8% | — |
| LMArena WebDev | — | 1633 |
| BigCodeBench Instruct | 48.6% | — |
| LMArena Coding | 1310 | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Reasoning Hy4 preview leads
DeepSeek-V2.5 (Sep 2024): 24.5 (#145), Hy4 preview: 30.6 (#75)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Hy4 preview |
|---|---|---|
| NYT Connections (extended) | — | 68.2% |
| LMArena Hard Prompts | 1289 | — |
Math Hy4 preview leads
DeepSeek-V2.5 (Sep 2024): 33.3 (#177), Hy4 preview: 54.5 (#37)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Hy4 preview |
|---|---|---|
| ProofBench | — | 75% |
| LMArena Math | 1289 | — |
Knowledge Not comparable
DeepSeek-V2.5 (Sep 2024): 34.2 (#190), Hy4 preview: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Hy4 preview |
|---|---|---|
| LMArena Expert | 1266 | — |
Multilingual Not comparable
DeepSeek-V2.5 (Sep 2024): 42.7 (#188), Hy4 preview: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Hy4 preview |
|---|---|---|
| LMArena Non-English | 1273 | — |
| LMArena Chinese | 1317 | — |
| LMArena French | 1288 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
| LMArena Russian | 1289 | — |
| LMArena Spanish | 1248 | — |
Instruction Following Not comparable
DeepSeek-V2.5 (Sep 2024): 67.7 (#189), Hy4 preview: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Hy4 preview |
|---|---|---|
| LMArena Instruction Following | 1280 | — |
Long Context Not comparable
DeepSeek-V2.5 (Sep 2024): 39.5 (#169), Hy4 preview: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Hy4 preview |
|---|---|---|
| LMArena Longer Query | 1301 | — |
Writing & Preference Not comparable
DeepSeek-V2.5 (Sep 2024): 50.0 (#182), Hy4 preview: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Hy4 preview |
|---|---|---|
| LMArena Text | 1294 | — |
| LMArena Creative Writing | 1285 | — |
| LMArena Multi-Turn | 1297 | — |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Hy4 preview?
Hy4 preview is the stronger model overall, scoring 44.6 to 37.0 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Hy4 preview better for coding?
Hy4 preview scores higher on coding benchmarks: 51.4 versus 31.3 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Hy4 preview share?
0 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Hy4 preview has 3.