Model comparison
DeepSeek V4.1 Flash vs Hy3
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 44.2 on the Noometry Index. Hy3 costs 1.8× less per token, which makes it the better buy when DeepSeek V4.1 Flash's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
DeepSeek
TencentSummary
- They share 19 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 8 categories and Hy3 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 40.1.
- The biggest single-benchmark swing is NYT Connections (extended): 89.6% for DeepSeek V4.1 Flash and 41.2% for Hy3.
- Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $0.15 / $0.60 for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 262K.
Side by side
| DeepSeek V4.1 Flash | Hy3 | |
|---|---|---|
| Provider | DeepSeek | Tencent |
| Noometry Index | 52.8 | 44.2 |
| Released | 2026-09-09 | 2026-07-06 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 393K | 128K |
| Input $ / M tokens | $0.15 | $0.0825 |
| Output $ / M tokens | $0.60 | $0.33 |
| Results tracked | 37 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 52.9 (#32), Hy3: 46.8 (#63)
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| LMArena WebDev | 1619 | 1508 |
| LMArena Coding | 1506 | 1464 |
| SciCode | 51.9% | — |
| ALE-Bench | 1,092 | — |
Agentic & Tool Use Not comparable
DeepSeek V4.1 Flash: 31.2 (#69), Hy3: —
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| APEX-Agents | 39.5% | — |
| GDP.pdf | 19.8% | — |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), Hy3: 26.1 (#136)
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| NYT Connections (extended) | 89.6% | 41.2% |
| LMArena Hard Prompts | 1483 | 1447 |
| CritPt | 14.3% | — |
| Mystery Game Puzzles | 43% | — |
| DTBench | 89.9% | — |
| LMCA | 47% | — |
| Surface Evolver Bench | 46.3% | — |
| Epoch Capabilities Index | 154.9 | — |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), Hy3: 40.1 (#93)
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| LMArena Math | 1477 | 1475 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 26.8% | — |
| OTIS Mock AIME 2024-2025 | 98.3% | — |
| ProofBench | 54% | — |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), Hy3: 40.8 (#114)
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| LMArena Expert | 1506 | 1460 |
| GPQA Diamond | 89.8% | — |
Multimodal Not comparable
DeepSeek V4.1 Flash: 39.1 (#61), Hy3: —
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| LMArena Vision | 1277 | — |
| Furniture Assembly | 34.2% | — |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), Hy3: 53.5 (#65)
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| LMArena Non-English | 1448 | 1426 |
| LMArena Chinese | 1497 | 1493 |
| LMArena French | 1452 | 1461 |
| LMArena German | 1484 | 1439 |
| LMArena Japanese | 1412 | 1392 |
| LMArena Korean | 1452 | 1395 |
| LMArena Russian | 1471 | 1432 |
| LMArena Spanish | 1459 | 1456 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.3 (#26), Hy3: 75.1 (#70)
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| LMArena Instruction Following | 1474 | 1426 |
Long Context DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 45.2 (#47), Hy3: 44.1 (#75)
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| LMArena Longer Query | 1475 | 1442 |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 65.4 (#48), Hy3: 62.2 (#81)
| Benchmark | DeepSeek V4.1 Flash | Hy3 |
|---|---|---|
| LMArena Text | 1462 | 1439 |
| LMArena Creative Writing | 1435 | 1402 |
| LMArena Multi-Turn | 1457 | 1436 |
| EQ-Bench Creative Writing | 1540 | — |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Hy3?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 44.2 on the Noometry Index. Hy3 costs 1.8× less per token, which makes it the better buy when DeepSeek V4.1 Flash's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4.1 Flash or Hy3?
Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; DeepSeek V4.1 Flash lists at $0.15 and $0.60.
Is DeepSeek V4.1 Flash or Hy3 better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 46.8 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4.1 Flash does, with 1M tokens against 262K.
How many benchmarks do DeepSeek V4.1 Flash and Hy3 share?
19 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Hy3 has 19.