DeepSeek V4 Pro 0423 vs Step 5 Preview
Step 5 Preview leads DeepSeek V4 Pro 0423 on coding, 61.1 to 38.7 on the Vibe Coding Index, at about the same output price ($2.70 vs $2.976 per million). Updated October 22, 2026; 38 benchmarks have a result for both.
DeepSeek V4 Pro 0423 or Step 5 Preview: which to pick
Pick Step 5 Preview for coding: it scores 61.1 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423, though on 4 benchmarks against 5.
Pick DeepSeek V4 Pro 0423 for work measured by LMArena Text: hard prompts in English or LMArena Text: science, where it leads.
- Shared results
- 38
- DeepSeek V4 Pro 0423 only
- 46
- Step 5 Preview only
- 23
- Like-for-like categories
- 7 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
Step 5 Preview
Step 5 Preview scores 61.1 against 38.7.
5 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
Step 5 Preview
Step 5 Preview leads, 43.7 against 30.4.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only DeepSeek V4 Pro 0423 has a AA Agentic Index result, 26.3.
Directional only
Reasoning
GPQA Diamond
No call
Only DeepSeek V4 Pro 0423 has a GPQA Diamond result, 89.4%.
Directional only
Knowledge
MMLU Pro
No call
Only DeepSeek V4 Pro 0423 has a MMLU Pro result, 87.3%.
Directional only
Web design
Design Arena: websites
No call
Only DeepSeek V4 Pro 0423 has a Design Arena: websites result, 1234.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Step 5 Preview
DeepSeek V4 Pro 0423 $0.0028, Step 5 Preview $0.0024 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only DeepSeek V4 Pro 0423 has the number, $0.57.
Directional only
At a glance
- Coding index
- DeepSeek V4 Pro 0423 38.7 from 5 benchmarks, Step 5 Preview 61.1 from 4 benchmarks.
- Widest category
- Coding, 15.6 points apart on average over 8 shared benchmarks.
- Widest single gap
- Terminal-Bench 4 (Vals AI): DeepSeek V4 Pro 0423 11.1% against Step 5 Preview 31.8%, 27.6 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Pro 0423 $2.976 against Step 5 Preview $2.70 per million output tokens, 1.1x 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 Pro 0423 | Bars | Step 5 Preview | Ties |
|---|---|---|---|---|
| Coding | 0 | 8 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 4 | ||
| Reasoning | 1 | 2 | ||
| Math | 0 | 3 | ||
| Knowledge | 4 | 0 | ||
| General | 14 | 1 | ||
| All shared boards | 19 | 19 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index08
| Benchmark | DeepSeek V4 Pro 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 85.6% | Only Step 5 Preview measured |
| LMArena WebDev | 1447 | 1567 | Step 5 Preview |
| AA Coding Index | 59.4 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LiveCodeBench (Vals AI) | 87.5% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| SWE-bench Verified (Vals AI) | 77.4% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Terminal-Bench 4 (Vals AI) | 11.1% | 31.8% | Step 5 Preview |
| Vibe Code Bench (Vals AI) | 49.9% | 70.4% | Step 5 Preview |
| Code migration (Vals AI) | 26.2% | 39.0% | Step 5 Preview |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| IOI (Vals AI) | Not measured | 38.1% | Only Step 5 Preview measured |
| Terminal-Bench 2.1 (Vals AI) | 50.2% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 40.8% | Only Step 5 Preview measured |
| SkillsBench (Vals AI) | 51.3% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| SciCode | Not measured | 58.9% | Only Step 5 Preview measured |
| FrontierCode | 17.6% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| WeirdML | 48.9% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| ALE-Bench | 1006.1 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: coding | 1470 | 1485 | Step 5 Preview |
| LMArena Text: software and IT | 1469 | 1473 | Step 5 Preview |
| LMArena WebDev: React | 1441 | 1583 | Step 5 Preview |
| LMArena WebDev: HTML | 1435 | 1556 | Step 5 Preview |
| LMArena WebDev: from an image | Not measured | 1550 | Only Step 5 Preview measured |
| Neither model has a result on the other 17 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | DeepSeek V4 Pro 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| AA Intelligence Index | 30.4 | 43.7 | Step 5 Preview |
| Epoch Capabilities Index | 149.1 | Not measured | Only DeepSeek V4 Pro 0423 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | DeepSeek V4 Pro 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| AA Agentic Index | 26.3 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: agents building web apps | 1000 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: agents building full-stack apps | 948 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: agents building Godot games | 1059 | Not measured | Only DeepSeek V4 Pro 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) | 44.1% | 50.7% | Step 5 Preview |
| Tax Agent Bench (Vals AI) | 25.2% | 27.3% | Step 5 Preview |
| Legal Research Bench (Vals AI) | 23.1% | 40.4% | Step 5 Preview |
| Excel Modeling Benchmark (Vals AI) | 51.6% | 57.1% | Step 5 Preview |
| 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 |
| Vending-Bench 2 | $3,285 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 25 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index12
| Benchmark | DeepSeek V4 Pro 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| GPQA Diamond | 89.4% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| 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 |
| GPQA Diamond (Epoch AI run) | 90.9% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Chess puzzles | 19.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Mystery game puzzles | 17.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: hard prompts | 1461 | 1461 | Step 5 Preview |
| LMArena Text: expert | 1463 | 1485 | Step 5 Preview |
| LMArena Text: hard prompts in English | 1466 | 1454 | DeepSeek V4 Pro 0423 |
| LMArena Vision: diagrams | Not measured | 1308 | Only Step 5 Preview measured |
| LMArena Vision: homework | Not measured | 1265 | Only Step 5 Preview measured |
| Kagi LLM Benchmark | 53.5% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 9 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | DeepSeek V4 Pro 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| OTIS Mock AIME | 96.7% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| FrontierMath, tiers 1 to 3 | 45.3% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| FrontierMath, tier 4 | 2.4% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| ProofBench | 16.0% | 42.0% | Step 5 Preview |
| LMArena Text: math | 1443 | 1458 | Step 5 Preview |
| LMArena Text: mathematical work | 1455 | 1460 | Step 5 Preview |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | DeepSeek V4 Pro 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| MMLU Pro | 87.3% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LegalBench (Vals AI) | 80.3% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| CaseLaw (Vals AI) | 59.4% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| CorpFin (Vals AI) | 61.4% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| TaxEval (Vals AI) | 72.1% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| MedCode (Vals AI) | 40.5% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| SimpleQA Verified | 47.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: science | 1470 | 1454 | DeepSeek V4 Pro 0423 |
| LMArena Text: medicine and healthcare | 1466 | 1439 | DeepSeek V4 Pro 0423 |
| LMArena Text: legal and government | 1464 | 1437 | DeepSeek V4 Pro 0423 |
| LMArena Text: business and finance | 1443 | 1435 | DeepSeek V4 Pro 0423 |
| Neither model has a result on the other 7 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | DeepSeek V4 Pro 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| Design Arena: websites | 1234 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: UI components | 1219 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: game dev | 1228 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: data viz | 1203 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: 3D | 1253 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: all code categories | 1240 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: SVG | 1139 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: ASCII art | 1149 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LiveBench design | Not measured | 76.9% | Only Step 5 Preview measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index141
| Benchmark | DeepSeek V4 Pro 0423 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 63.5% | Only Step 5 Preview measured |
| Public Benefits Bench (Vals AI) | 62.9% | 58.1% | DeepSeek V4 Pro 0423 |
| MedScribe (Vals AI) | 75.1% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Long context reasoning (AA-LCR) | Not measured | 88.3% | Only Step 5 Preview measured |
| DTBench | 90.7% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMCA | 41.2% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Surface Evolver Bench | 40.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| CL-bench Life | 13.5% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text | 1451 | 1440 | DeepSeek V4 Pro 0423 |
| LMArena Text: instruction following | 1445 | 1438 | DeepSeek V4 Pro 0423 |
| LMArena Text: creative writing | 1446 | 1409 | DeepSeek V4 Pro 0423 |
| LMArena Vision | Not measured | 1265 | Only Step 5 Preview measured |
| LMArena Text: English | 1459 | 1447 | DeepSeek V4 Pro 0423 |
| LMArena Text: non-English | 1439 | 1428 | DeepSeek V4 Pro 0423 |
| LMArena Text: Chinese | 1481 | 1500 | Step 5 Preview |
| LMArena Text: German | 1458 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: Japanese | 1446 | 1383 | DeepSeek V4 Pro 0423 |
| LMArena Text: Korean | 1409 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: French | 1473 | 1438 | DeepSeek V4 Pro 0423 |
| LMArena Text: Spanish | 1457 | 1441 | DeepSeek V4 Pro 0423 |
| LMArena Text: Russian | 1453 | 1433 | DeepSeek V4 Pro 0423 |
| LMArena Text: Polish | 1461 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: multi-turn | 1468 | 1437 | DeepSeek V4 Pro 0423 |
| LMArena Text: longer queries | 1458 | 1449 | DeepSeek V4 Pro 0423 |
| LMArena Text: writing and language | 1447 | 1425 | DeepSeek V4 Pro 0423 |
| LMArena Text: entertainment and media | 1431 | 1412 | DeepSeek V4 Pro 0423 |
| 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 18 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Pro 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
Step 5 Preview costs less
Repository review
50K input + 3K output tokens
Step 5 Preview costs less
Agent loop
200K input + 10K output tokens
Step 5 Preview 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 Pro 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 | $1.358 | $1.00 |
| Output $/Mtok | $2.976 | $2.70 |
| Coding index | 38.7 | 61.1 |
| Independent measurements behind it | 5 | 4 |
| Cost per solved task | $0.57 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Pro 0423 better than Step 5 Preview for coding?
Pick Step 5 Preview for coding: it scores 61.1 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423, though on 4 benchmarks against 5. Pick DeepSeek V4 Pro 0423 for work measured by LMArena Text: hard prompts in English or LMArena Text: science, where it leads.
Which is cheaper, DeepSeek V4 Pro 0423 or Step 5 Preview?
DeepSeek V4 Pro 0423 costs $1.358 per million input tokens and $2.976 per million output tokens; Step 5 Preview costs $1.00 and $2.70. Cost per solved task is known for DeepSeek V4 Pro 0423 only.
Which has the bigger context window, DeepSeek V4 Pro 0423 or Step 5 Preview?
Neither: both take 1.05M tokens.
How many benchmarks is the DeepSeek V4 Pro 0423 vs Step 5 Preview comparison based on?
38 benchmarks have a result for both models as of October 22, 2026. DeepSeek V4 Pro 0423's coding index rests on 5 independent benchmarks and Step 5 Preview's on 4. Where only one model has a result, the page says so instead of naming a winner.