DeepSeek V4 Flash 0423 vs Hy3
DeepSeek V4 Flash 0423 costs 1.9x less per output token than Hy3 ($0.28 vs $0.528 per million). Updated October 22, 2026; 40 benchmarks have a result for both.
DeepSeek V4 Flash 0423 or Hy3: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick DeepSeek V4 Flash 0423 for output tokens 1.9x cheaper ($0.28 vs $0.528 per million) and a larger context window (1.05M against 262K tokens).
- Shared results
- 40
- DeepSeek V4 Flash 0423 only
- 9
- Hy3 only
- 4
- Like-for-like categories
- 6 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
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
No call
Only DeepSeek V4 Flash 0423 has a AA Intelligence Index result, 24.4.
Directional only
Agentic work
AA Agentic Index
No call
Only DeepSeek V4 Flash 0423 has a AA Agentic Index result, 26.3.
Directional only
Web design
Design Arena: websites
Hy3
Hy3 leads, 1230 against 1207.
Same board, both measured
Long documents
Listed context window
DeepSeek V4 Flash 0423
DeepSeek V4 Flash 0423 takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Flash 0423
DeepSeek V4 Flash 0423 $0.0003, Hy3 $0.0004 per turn.
List prices
At a glance
- Coding index
- DeepSeek V4 Flash 0423 not benchmarked, Hy3 not benchmarked.
- Widest category
- Math, 8.1 points apart on average over 2 shared benchmarks.
- Widest single gap
- LMArena WebDev: React: DeepSeek V4 Flash 0423 1585 against Hy3 1511, 12.3 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Flash 0423 $0.28 against Hy3 $0.528 per million output tokens, 1.9x apart.
- Context window
- DeepSeek V4 Flash 0423 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 0423 | Bars | Hy3 | Ties |
|---|---|---|---|---|
| Coding | 3 | 3 | ||
| Reasoning | 0 | 3 | ||
| Math | 0 | 2 | ||
| Knowledge | 2 | 2 | ||
| Web design | 0 | 8 | ||
| General | 3 | 14 | ||
| All shared boards | 8 | 32 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index33
| Benchmark | DeepSeek V4 Flash 0423 | Hy3 | Leader |
|---|---|---|---|
| LMArena WebDev | 1582 | 1507 | DeepSeek V4 Flash 0423 |
| AA Coding Index | 52.0 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| SciCode | Not measured | 48.6% | Only Hy3 measured |
| WeirdML | 45.6% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| ALE-Bench | 678.2 | 838.6 | Hy3 |
| LMArena Text: coding | 1457 | 1466 | Hy3 |
| LMArena Text: software and IT | 1453 | 1463 | Hy3 |
| LMArena WebDev: React | 1585 | 1511 | DeepSeek V4 Flash 0423 |
| LMArena WebDev: HTML | 1558 | 1487 | DeepSeek V4 Flash 0423 |
| Neither model has a result on the other 30 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Hy3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 24.4 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Epoch Capabilities Index | 146.1 | Not measured | Only DeepSeek V4 Flash 0423 measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Hy3 | Leader |
|---|---|---|---|
| AA Agentic Index | 26.3 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Neither model has a result on the other 40 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | DeepSeek V4 Flash 0423 | Hy3 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 89.7% | Only Hy3 measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 33.5% | Only Hy3 measured |
| SimpleBench | 46.3% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: hard prompts | 1444 | 1447 | Hy3 |
| LMArena Text: expert | 1441 | 1461 | Hy3 |
| LMArena Text: hard prompts in English | 1449 | 1455 | Hy3 |
| Kagi LLM Benchmark | 52.2% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Neither model has a result on the other 15 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V4 Flash 0423 | Hy3 | Leader |
|---|---|---|---|
| LMArena Text: math | 1426 | 1473 | Hy3 |
| LMArena Text: mathematical work | 1432 | 1482 | Hy3 |
| Neither model has a result on the other 11 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | DeepSeek V4 Flash 0423 | Hy3 | Leader |
|---|---|---|---|
| LMArena Text: science | 1446 | 1457 | Hy3 |
| LMArena Text: medicine and healthcare | 1448 | 1452 | Hy3 |
| LMArena Text: legal and government | 1439 | 1422 | DeepSeek V4 Flash 0423 |
| LMArena Text: business and finance | 1425 | 1424 | DeepSeek V4 Flash 0423 |
| Neither model has a result on the other 14 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | DeepSeek V4 Flash 0423 | Hy3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1207 | 1230 | Hy3 |
| Design Arena: UI components | 1162 | 1204 | Hy3 |
| Design Arena: game dev | 1196 | 1208 | Hy3 |
| Design Arena: data viz | 1127 | 1164 | Hy3 |
| Design Arena: 3D | 1188 | 1232 | Hy3 |
| Design Arena: all code categories | 1205 | 1227 | Hy3 |
| Design Arena: SVG | 1155 | 1200 | Hy3 |
| Design Arena: ASCII art | 1109 | 1152 | Hy3 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index314
| Benchmark | DeepSeek V4 Flash 0423 | Hy3 | Leader |
|---|---|---|---|
| Long context reasoning (AA-LCR) | Not measured | 79.0% | Only Hy3 measured |
| DTBench | 86.4% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMCA | 35.9% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text | 1432 | 1438 | Hy3 |
| LMArena Text: instruction following | 1421 | 1426 | Hy3 |
| LMArena Text: creative writing | 1403 | 1403 | DeepSeek V4 Flash 0423 |
| LMArena Text: English | 1441 | 1448 | Hy3 |
| LMArena Text: non-English | 1419 | 1426 | Hy3 |
| LMArena Text: Chinese | 1463 | 1485 | Hy3 |
| LMArena Text: German | 1419 | 1449 | Hy3 |
| LMArena Text: Japanese | 1389 | 1397 | Hy3 |
| LMArena Text: Korean | 1383 | 1396 | Hy3 |
| LMArena Text: French | 1439 | 1456 | Hy3 |
| LMArena Text: Spanish | 1436 | 1457 | Hy3 |
| LMArena Text: Russian | 1427 | 1432 | Hy3 |
| LMArena Text: Polish | 1436 | 1447 | Hy3 |
| LMArena Text: multi-turn | 1449 | 1436 | DeepSeek V4 Flash 0423 |
| LMArena Text: longer queries | 1434 | 1442 | Hy3 |
| LMArena Text: writing and language | 1410 | 1414 | Hy3 |
| LMArena Text: entertainment and media | 1400 | 1400 | DeepSeek V4 Flash 0423 |
| Neither model has a result on the other 26 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | 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
DeepSeek V4 Flash 0423 costs less
Repository review
50K input + 3K output tokens
DeepSeek V4 Flash 0423 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4 Flash 0423 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 0423 | Tencent Hy3 |
|---|---|---|
| Maker | DeepSeek | Tencent |
| Weights | Open weights | Open weights |
| Released | April 24, 2026 | July 6, 2026 |
| Context window | 1.05M | 262K |
| Input $/Mtok | $0.14 | $0.132 |
| Output $/Mtok | $0.28 | $0.528 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 1 | 1 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Flash 0423 better than Hy3 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick DeepSeek V4 Flash 0423 for output tokens 1.9x cheaper ($0.28 vs $0.528 per million) and a larger context window (1.05M against 262K tokens).
Which is cheaper, DeepSeek V4 Flash 0423 or Hy3?
DeepSeek V4 Flash 0423 costs $0.14 per million input tokens and $0.28 per million output tokens; Hy3 costs $0.132 and $0.528. Cost per solved task is known for neither of them.
Which has the bigger context window, DeepSeek V4 Flash 0423 or Hy3?
DeepSeek V4 Flash 0423, with 1.05M tokens against 262K.
How many benchmarks is the DeepSeek V4 Flash 0423 vs Hy3 comparison based on?
40 benchmarks have a result for both models as of October 22, 2026. DeepSeek V4 Flash 0423's coding index rests on 1 independent benchmark and Hy3's on 1. Where only one model has a result, the page says so instead of naming a winner.