DeepSeek V4 Flash 0423 vs Step 5 Preview
Updated October 21, 2026. 29 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick Step 5 Preview for coding: it scores 71.4 on the Vibe Coding Index against 54.6 for DeepSeek V4 Flash 0423.
Pick DeepSeek V4 Flash 0423 when output tokens cost 14.8x less, or when what matters is LMArena WebDev or LMArena WebDev: React, where it leads.
- Shared results
- 29
- DeepSeek V4 Flash 0423 only
- 19
- Step 5 Preview only
- 32
- Like-for-like categories
- 6 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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, four public coding benchmarks
Step 5 Preview
Step 5 Preview scores 71.4 against 54.6.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
Step 5 Preview
Step 5 Preview leads, 43.7 against 24.4.
Same board, both measured
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
No call
Only DeepSeek V4 Flash 0423 has a Design Arena: websites result, 1208.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Flash 0423
DeepSeek V4 Flash 0423 $0.0002, Step 5 Preview $0.0024 per turn.
List prices
At a glance
- Coding index
- DeepSeek V4 Flash 0423 54.6 from 1 benchmark, Step 5 Preview 71.4 from 2 benchmarks.
- Widest category
- Intelligence, 19.3 points apart on average over 1 shared benchmark.
- Widest single gap
- AA Intelligence Index: DeepSeek V4 Flash 0423 24.4 against Step 5 Preview 43.7, 19.3 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Flash 0423 $0.182 against Step 5 Preview $2.70 per million output tokens, 14.8x apart.
- Context window
- Both take 1.05M tokens.
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 | Step 5 Preview | Ties |
|---|---|---|---|---|
| Coding | 3 | 2 | ||
| Intelligence | 0 | 1 | ||
| Reasoning | 0 | 3 | ||
| Math | 0 | 2 | ||
| Knowledge | 2 | 2 | ||
| General | 3 | 11 | ||
| All shared boards | 8 | 21 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index32
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 85.6% | Only Step 5 Preview measured |
| LMArena WebDev | 1582 | 1567 | DeepSeek V4 Flash 0423 |
| AA Coding Index | 52.0 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 31.8% | Only Step 5 Preview measured |
| Vibe Code Bench (Vals AI) | Not measured | 70.4% | Only Step 5 Preview measured |
| Code migration (Vals AI) | Not measured | 39.0% | Only Step 5 Preview measured |
| IOI (Vals AI) | Not measured | 38.1% | Only Step 5 Preview measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 40.8% | Only Step 5 Preview measured |
| SciCode | Not measured | 58.9% | Only Step 5 Preview measured |
| WeirdML | 45.6% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| ALE-Bench | 678.2 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: coding | 1457 | 1485 | Step 5 Preview |
| LMArena Text: software and IT | 1453 | 1473 | Step 5 Preview |
| LMArena WebDev: React | 1585 | 1583 | DeepSeek V4 Flash 0423 |
| LMArena WebDev: HTML | 1558 | 1556 | DeepSeek V4 Flash 0423 |
| LMArena WebDev: from an image | Not measured | 1550 | Only Step 5 Preview measured |
| Neither model has a result on the other 22 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| AA Intelligence Index | 24.4 | 43.7 | Step 5 Preview |
| Epoch Capabilities Index | 146.1 | Not measured | Only DeepSeek V4 Flash 0423 measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index12 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| AA Agentic Index | 26.3 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LiveBench research | Not measured | 72.8% | Only Step 5 Preview measured |
| LiveBench browser | Not measured | 24.0% | Only Step 5 Preview measured |
| LiveBench long agentic | Not measured | 27.5% | Only Step 5 Preview measured |
| SREBench (Vals AI) | Not measured | 0.4% | Only Step 5 Preview measured |
| Finance Agent (Vals AI) | Not measured | 50.7% | Only Step 5 Preview measured |
| Tax Agent Bench (Vals AI) | Not measured | 27.3% | Only Step 5 Preview measured |
| Legal Research Bench (Vals AI) | Not measured | 40.4% | Only Step 5 Preview measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 57.1% | Only Step 5 Preview measured |
| BioMysteryBench (Vals AI) | Not measured | 67.4% | Only Step 5 Preview measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 2.9% | Only Step 5 Preview measured |
| Vals Science Index | Not measured | 29.6% | Only Step 5 Preview measured |
| Neither model has a result on the other 27 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench reasoning | Not measured | 83.8% | Only Step 5 Preview measured |
| MysteryMechanism (Vals AI) | Not measured | 18.5% | Only Step 5 Preview measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 46.5% | Only Step 5 Preview measured |
| SimpleBench | 46.3% | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: hard prompts | 1444 | 1461 | Step 5 Preview |
| LMArena Text: expert | 1441 | 1485 | Step 5 Preview |
| LMArena Text: hard prompts in English | 1449 | 1454 | Step 5 Preview |
| LMArena Vision: diagrams | Not measured | 1308 | Only Step 5 Preview measured |
| LMArena Vision: homework | Not measured | 1265 | Only Step 5 Preview measured |
| Neither model has a result on the other 12 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| ProofBench | Not measured | 42.0% | Only Step 5 Preview measured |
| LMArena Text: math | 1426 | 1458 | Step 5 Preview |
| LMArena Text: mathematical work | 1432 | 1460 | Step 5 Preview |
| Neither model has a result on the other 10 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| LMArena Text: science | 1446 | 1454 | Step 5 Preview |
| LMArena Text: medicine and healthcare | 1448 | 1439 | DeepSeek V4 Flash 0423 |
| LMArena Text: legal and government | 1439 | 1437 | DeepSeek V4 Flash 0423 |
| LMArena Text: business and finance | 1425 | 1435 | Step 5 Preview |
| Neither model has a result on the other 13 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| Design Arena: websites | 1208 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: UI components | 1164 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: game dev | 1197 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: data viz | 1127 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: 3D | 1189 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: all code categories | 1205 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: SVG | 1155 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| Design Arena: ASCII art | 1109 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LiveBench design | Not measured | 76.9% | Only Step 5 Preview measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index311
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 63.5% | Only Step 5 Preview measured |
| Public Benefits Bench (Vals AI) | Not measured | 58.1% | Only Step 5 Preview measured |
| Long context reasoning (AA-LCR) | Not measured | 88.3% | Only Step 5 Preview 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 | 1440 | Step 5 Preview |
| LMArena Text: instruction following | 1421 | 1438 | Step 5 Preview |
| LMArena Text: creative writing | 1403 | 1409 | Step 5 Preview |
| LMArena Vision | Not measured | 1265 | Only Step 5 Preview measured |
| LMArena Text: English | 1441 | 1447 | Step 5 Preview |
| LMArena Text: non-English | 1419 | 1428 | Step 5 Preview |
| LMArena Text: Chinese | 1463 | 1500 | Step 5 Preview |
| LMArena Text: German | 1419 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: Japanese | 1389 | 1383 | DeepSeek V4 Flash 0423 |
| LMArena Text: Korean | 1383 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: French | 1439 | 1438 | DeepSeek V4 Flash 0423 |
| LMArena Text: Spanish | 1436 | 1441 | Step 5 Preview |
| LMArena Text: Russian | 1427 | 1433 | Step 5 Preview |
| LMArena Text: Polish | 1436 | Not measured | Only DeepSeek V4 Flash 0423 measured |
| LMArena Text: multi-turn | 1449 | 1437 | DeepSeek V4 Flash 0423 |
| LMArena Text: longer queries | 1434 | 1449 | Step 5 Preview |
| LMArena Text: writing and language | 1410 | 1425 | Step 5 Preview |
| LMArena Text: entertainment and media | 1400 | 1412 | Step 5 Preview |
| LMArena Vision: English | Not measured | 1290 | Only Step 5 Preview measured |
| LMArena Vision: reading text | Not measured | 1282 | Only Step 5 Preview measured |
| Neither model has a result on the other 21 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Flash 0423 | Step 5 Preview | 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 | StepFun Step 5 Preview |
|---|---|---|
| Maker | DeepSeek | StepFun |
| Weights | Open weights | Open weights |
| Released | April 24, 2026 | October 8, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.091 | $1.00 |
| Output $/Mtok | $0.182 | $2.70 |
| Coding index | 54.6 | 71.4 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Flash 0423 better than Step 5 Preview for coding?
Pick Step 5 Preview for coding: it scores 71.4 on the Vibe Coding Index against 54.6 for DeepSeek V4 Flash 0423. Pick DeepSeek V4 Flash 0423 when output tokens cost 14.8x less, or when what matters is LMArena WebDev or LMArena WebDev: React, where it leads.
Which is cheaper, DeepSeek V4 Flash 0423 or Step 5 Preview?
DeepSeek V4 Flash 0423 costs $0.091 per million input tokens and $0.182 per million output tokens; Step 5 Preview costs $1.00 and $2.70. Cost per solved task is known for neither of them.
Which has the bigger context window, DeepSeek V4 Flash 0423 or Step 5 Preview?
Neither: both take 1.05M tokens.
How many benchmarks is the DeepSeek V4 Flash 0423 vs Step 5 Preview comparison based on?
29 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Flash 0423's coding index rests on 1 independent benchmark and Step 5 Preview's on 2. Where only one model has a result, the page says so instead of naming a winner.