DeepSeek V4 Flash 0731 vs Hy3
DeepSeek V4 Flash 0731 and Hy3 cost about the same per output token ($0.28 vs $0.33 per million). Updated October 21, 2026; 13 benchmarks have a result for both.
DeepSeek V4 Flash 0731 or Hy3: which to pick
Only DeepSeek V4 Flash 0731 has a coding score, 51.8 on the Vibe Coding Index; Hy3 has none, so the two cannot be ranked on coding.
Pick Hy3 for work measured by Design Arena: 3D or Design Arena: SVG, where it leads.
- Shared results
- 13
- DeepSeek V4 Flash 0731 only
- 42
- Hy3 only
- 31
- Like-for-like categories
- 4 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Only DeepSeek V4 Flash 0731 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
No call
Only DeepSeek V4 Flash 0731 has a AA Intelligence Index result, 34.3.
Directional only
Agentic work
AA Agentic Index
No call
Only DeepSeek V4 Flash 0731 has a AA Agentic Index result, 41.0.
Directional only
Reasoning
GPQA Diamond
No call
Only DeepSeek V4 Flash 0731 has a GPQA Diamond result, 89.9%.
Directional only
Knowledge
MMLU Pro
No call
Only DeepSeek V4 Flash 0731 has a MMLU Pro result, 86.2%.
Directional only
Web design
Design Arena: websites
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 1238 against 1230.
Same board, both measured
Long documents
Listed context window
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Hy3
DeepSeek V4 Flash 0731 $0.0003, Hy3 $0.0002 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only DeepSeek V4 Flash 0731 has the number, $0.01.
Directional only
At a glance
- Coding index
- DeepSeek V4 Flash 0731 51.8 from 4 benchmarks, Hy3 not benchmarked.
- Widest category
- Coding, 8.6 points apart on average over 2 shared benchmarks.
- Widest single gap
- ALE-Bench: DeepSeek V4 Flash 0731 1306.1 against Hy3 838.6, 15.6 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Flash 0731 $0.28 against Hy3 $0.33 per million output tokens, 1.2x apart.
- Context window
- DeepSeek V4 Flash 0731 takes 1.05M tokens, Hy3 262K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | DeepSeek V4 Flash 0731 | Bars | Hy3 | Ties |
|---|---|---|---|---|
| Coding | 2 | 0 | ||
| Reasoning | 2 | 0 | ||
| Web design | 4 | 3 | 1 tie | |
| General | 1 | 0 | ||
| All shared boards | 9 | 3 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index20
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| LMArena WebDev | Not measured | 1507 | Only Hy3 measured |
| AA Coding Index | 69.1 | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LiveCodeBench (Vals AI) | 87.3% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SWE-bench Verified (Vals AI) | 88.8% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench 4 (Vals AI) | 18.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Vibe Code Bench (Vals AI) | 74.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Code migration (Vals AI) | 38.6% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| IOI (Vals AI) | 32.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench 2.1 (Vals AI) | 67.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SkillsBench (Vals AI) | 50.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| CyberBench (Vals AI) | 72.1% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SciCode | 50.3% | 48.6% | DeepSeek V4 Flash 0731 |
| FrontierCode | 18.8% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| WeirdML | 63.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ALE-Bench | 1306.1 | 838.6 | DeepSeek V4 Flash 0731 |
| LMArena Text: coding | Not measured | 1466 | Only Hy3 measured |
| LMArena Text: software and IT | Not measured | 1463 | Only Hy3 measured |
| LMArena WebDev: React | Not measured | 1511 | Only Hy3 measured |
| LMArena WebDev: HTML | Not measured | 1487 | Only Hy3 measured |
| Neither model has a result on the other 19 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 34.3 | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Epoch Capabilities Index | 154.8 | Not measured | Only DeepSeek V4 Flash 0731 measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| AA Agentic Index | 41.0 | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Finance Agent (Vals AI) | 49.5% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Tax Agent Bench (Vals AI) | 28.1% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Legal Research Bench (Vals AI) | 30.3% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Excel Modeling Benchmark (Vals AI) | 57.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| BioMysteryBench (Vals AI) | 64.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench-Science (Vals AI) | 4.3% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Neither model has a result on the other 34 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| GPQA Diamond | 89.9% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| GPQA Diamond (Artificial Analysis run) | 90.8% | 89.7% | DeepSeek V4 Flash 0731 |
| Humanity's Last Exam (Artificial Analysis run) | 38.6% | 33.5% | DeepSeek V4 Flash 0731 |
| GPQA Diamond (Epoch AI run) | 91.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ARC-AGI | 89.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ARC-AGI-2 | 61.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SimpleBench | 61.1% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Chess puzzles | 32.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Mystery game puzzles | 34.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LMArena Text: hard prompts | Not measured | 1447 | Only Hy3 measured |
| LMArena Text: expert | Not measured | 1461 | Only Hy3 measured |
| LMArena Text: hard prompts in English | Not measured | 1455 | Only Hy3 measured |
| Neither model has a result on the other 10 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 94.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| FrontierMath, tiers 1 to 3 | 57.5% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| FrontierMath, tier 4 | 24.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ProofBench | 56.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LMArena Text: math | Not measured | 1473 | Only Hy3 measured |
| LMArena Text: mathematical work | Not measured | 1482 | Only Hy3 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index10 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LegalBench (Vals AI) | 77.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| CorpFin (Vals AI) | 61.9% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| TaxEval (Vals AI) | 70.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| MedCode (Vals AI) | 41.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SimpleQA Verified | 33.6% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LMArena Text: science | Not measured | 1457 | Only Hy3 measured |
| LMArena Text: medicine and healthcare | Not measured | 1452 | Only Hy3 measured |
| LMArena Text: legal and government | Not measured | 1422 | Only Hy3 measured |
| LMArena Text: business and finance | Not measured | 1424 | Only Hy3 measured |
| Neither model has a result on the other 8 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index43
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1238 | 1230 | DeepSeek V4 Flash 0731 |
| Design Arena: UI components | 1235 | 1207 | DeepSeek V4 Flash 0731 |
| Design Arena: game dev | 1208 | 1208 | Tie |
| Design Arena: data viz | 1180 | 1163 | DeepSeek V4 Flash 0731 |
| Design Arena: 3D | 1209 | 1233 | Hy3 |
| Design Arena: all code categories | 1230 | 1227 | DeepSeek V4 Flash 0731 |
| Design Arena: SVG | 1185 | 1200 | Hy3 |
| Design Arena: ASCII art | 1090 | 1152 | Hy3 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| MedScribe (Vals AI) | 80.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Long context reasoning (AA-LCR) | 79.7% | 79.0% | DeepSeek V4 Flash 0731 |
| DTBench | 90.9% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LMCA | 41.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LMArena Text | Not measured | 1438 | Only Hy3 measured |
| LMArena Text: instruction following | Not measured | 1426 | Only Hy3 measured |
| LMArena Text: creative writing | Not measured | 1403 | Only Hy3 measured |
| LMArena Text: English | Not measured | 1448 | Only Hy3 measured |
| LMArena Text: non-English | Not measured | 1426 | Only Hy3 measured |
| LMArena Text: Chinese | Not measured | 1485 | Only Hy3 measured |
| LMArena Text: German | Not measured | 1449 | Only Hy3 measured |
| LMArena Text: Japanese | Not measured | 1397 | Only Hy3 measured |
| LMArena Text: Korean | Not measured | 1396 | Only Hy3 measured |
| LMArena Text: French | Not measured | 1456 | Only Hy3 measured |
| LMArena Text: Spanish | Not measured | 1457 | Only Hy3 measured |
| LMArena Text: Russian | Not measured | 1432 | Only Hy3 measured |
| LMArena Text: Polish | Not measured | 1447 | Only Hy3 measured |
| LMArena Text: multi-turn | Not measured | 1436 | Only Hy3 measured |
| LMArena Text: longer queries | Not measured | 1442 | Only Hy3 measured |
| LMArena Text: writing and language | Not measured | 1414 | Only Hy3 measured |
| LMArena Text: entertainment and media | Not measured | 1400 | Only Hy3 measured |
| Neither model has a result on the other 25 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | Hy3 | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Hy3 costs less
Repository review
50K input + 3K output tokens
Hy3 costs less
Agent loop
200K input + 10K output tokens
Hy3 costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | DeepSeek V4 Flash 0731 | Tencent Hy3 |
|---|---|---|
| Maker | DeepSeek | Tencent |
| Weights | Open weights | Open weights |
| Released | July 31, 2026 | July 6, 2026 |
| Context window | 1.05M | 262K |
| Input $/Mtok | $0.14 | $0.0825 |
| Output $/Mtok | $0.28 | $0.33 |
| Coding index | 51.8 | Not benchmarked |
| Independent measurements behind it | 4 | 1 |
| Cost per solved task | $0.01 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Flash 0731 better than Hy3 for coding?
Only DeepSeek V4 Flash 0731 has a coding score, 51.8 on the Vibe Coding Index; Hy3 has none, so the two cannot be ranked on coding. Pick Hy3 for work measured by Design Arena: 3D or Design Arena: SVG, where it leads.
Which is cheaper, DeepSeek V4 Flash 0731 or Hy3?
DeepSeek V4 Flash 0731 costs $0.14 per million input tokens and $0.28 per million output tokens; Hy3 costs $0.0825 and $0.33. Cost per solved task is known for DeepSeek V4 Flash 0731 only.
Which has the bigger context window, DeepSeek V4 Flash 0731 or Hy3?
DeepSeek V4 Flash 0731, with 1.05M tokens against 262K.
How many benchmarks is the DeepSeek V4 Flash 0731 vs Hy3 comparison based on?
13 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Flash 0731's coding index rests on 4 independent benchmarks and Hy3's on 1. Where only one model has a result, the page says so instead of naming a winner.