DeepSeek V4 Flash 0731 vs Grok 4.5
Grok 4.5 leads DeepSeek V4 Flash 0731 on coding, 58.6 to 51.8 on the Vibe Coding Index, and DeepSeek V4 Flash 0731 costs 21.4x less per output token ($0.28 vs $6.00 per million). Updated October 21, 2026; 52 benchmarks have a result for both.
DeepSeek V4 Flash 0731 or Grok 4.5: which to pick
Pick Grok 4.5 for coding: it scores 58.6 on the Vibe Coding Index against 51.8 for DeepSeek V4 Flash 0731.
Pick DeepSeek V4 Flash 0731 for output tokens 21.4x cheaper ($0.28 vs $6.00 per million) and a lower cost per solved task ($0.01 against $0.62).
- Shared results
- 52
- DeepSeek V4 Flash 0731 only
- 3
- Grok 4.5 only
- 71
- Like-for-like categories
- 8 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
Grok 4.5
Grok 4.5 scores 58.6 against 51.8.
4 and 6 benchmarks behind the scores
Intelligence
AA Intelligence Index
Grok 4.5
Grok 4.5 leads, 38.8 against 34.3.
Same board, both measured
Agentic work
AA Agentic Index
Grok 4.5
Grok 4.5 leads, 41.2 against 41.0.
Same board, both measured
Reasoning
GPQA Diamond
Grok 4.5
Grok 4.5 leads, 92.9% against 89.9%.
Same board, both measured
Knowledge
MMLU Pro
Grok 4.5
Grok 4.5 leads, 89.2% against 86.2%.
Same board, both measured
Web design
Design Arena: websites
Grok 4.5
Grok 4.5 leads, 1281 against 1238.
Same board, both measured
Long documents
Listed context window
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 takes 1.05M tokens against 500K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.0003, Grok 4.5 $0.0050 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.01, Grok 4.5 $0.62.
Published attempt cost and pass rate
At a glance
- Coding index
- DeepSeek V4 Flash 0731 51.8 from 4 benchmarks, Grok 4.5 58.6 from 6 benchmarks.
- Widest category
- Web design, 8.1 points apart on average over 8 shared benchmarks.
- Widest single gap
- FrontierCode: DeepSeek V4 Flash 0731 18.8% against Grok 4.5 42.4%, 36.4 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Flash 0731 $0.28 against Grok 4.5 $6.00 per million output tokens, 21.4x apart.
- Context window
- DeepSeek V4 Flash 0731 takes 1.05M tokens, Grok 4.5 500K.
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 | Grok 4.5 | Ties |
|---|---|---|---|---|
| Coding | 6 | 8 | 1 tie | |
| Intelligence | 1 | 1 | ||
| Agentic work | 3 | 2 | ||
| Reasoning | 2 | 6 | ||
| Math | 2 | 1 | 1 tie | |
| Knowledge | 0 | 6 | ||
| Web design | 0 | 8 | ||
| General | 1 | 3 | ||
| All shared boards | 15 | 35 | 2 ties |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index68
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 94.3% | Only Grok 4.5 measured |
| LMArena WebDev | Not measured | 1553 | Only Grok 4.5 measured |
| AA Coding Index | 69.1 | 72.4 | Grok 4.5 |
| LiveCodeBench (Vals AI) | 87.3% | 87.3% | Grok 4.5 |
| SWE-bench Verified (Vals AI) | 88.8% | 86.6% | DeepSeek V4 Flash 0731 |
| Terminal-Bench 4 (Vals AI) | 18.7% | 8.6% | DeepSeek V4 Flash 0731 |
| Vibe Code Bench (Vals AI) | 74.7% | 69.0% | DeepSeek V4 Flash 0731 |
| Code migration (Vals AI) | 38.6% | 36.6% | DeepSeek V4 Flash 0731 |
| ProgramBench (Vals AI) | 0.0% | 0.0% | Tie |
| IOI (Vals AI) | 32.7% | 40.6% | Grok 4.5 |
| Terminal-Bench 2.1 (Vals AI) | 67.0% | 67.8% | Grok 4.5 |
| SkillsBench (Vals AI) | 50.7% | 66.0% | Grok 4.5 |
| CyberBench (Vals AI) | 72.1% | 71.0% | DeepSeek V4 Flash 0731 |
| SciCode | 50.3% | 55.0% | Grok 4.5 |
| FrontierCode | 18.8% | 42.4% | Grok 4.5 |
| WeirdML | 63.0% | 46.4% | DeepSeek V4 Flash 0731 |
| ALE-Bench | 1306.1 | 1309.3 | Grok 4.5 |
| LMArena Text: coding | Not measured | 1474 | Only Grok 4.5 measured |
| LMArena Text: software and IT | Not measured | 1471 | Only Grok 4.5 measured |
| LMArena WebDev: React | Not measured | 1556 | Only Grok 4.5 measured |
| LMArena WebDev: HTML | Not measured | 1567 | Only Grok 4.5 measured |
| LMArena WebDev: from an image | Not measured | 1562 | Only Grok 4.5 measured |
| SciCode | Not measured | 6.5% | Only Grok 4.5 measured |
| Neither model has a result on the other 16 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | Leader |
|---|---|---|---|
| AA Intelligence Index | 34.3 | 38.8 | Grok 4.5 |
| Epoch Capabilities Index | 154.8 | 154.1 | DeepSeek V4 Flash 0731 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index32
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | Leader |
|---|---|---|---|
| AA Agentic Index | 41.0 | 41.2 | Grok 4.5 |
| Design Arena: agents building web apps | Not measured | 1218 | Only Grok 4.5 measured |
| Design Arena: agents building full-stack apps | Not measured | 1249 | Only Grok 4.5 measured |
| Design Arena: agents building mobile apps | Not measured | 1221 | Only Grok 4.5 measured |
| Design Arena: agents building Android apps | Not measured | 1242 | Only Grok 4.5 measured |
| Design Arena: agents building Godot games | Not measured | 1229 | Only Grok 4.5 measured |
| Design Arena: agents building games | Not measured | 1236 | Only Grok 4.5 measured |
| Design Arena: HTML slides | Not measured | 1204 | Only Grok 4.5 measured |
| Design Arena: PowerPoint slides | Not measured | 1233 | Only Grok 4.5 measured |
| LiveBench research | Not measured | 85.9% | Only Grok 4.5 measured |
| LiveBench browser | Not measured | 44.0% | Only Grok 4.5 measured |
| LiveBench long agentic | Not measured | 27.9% | Only Grok 4.5 measured |
| SREBench (Vals AI) | Not measured | 0.8% | Only Grok 4.5 measured |
| Finance Agent (Vals AI) | 49.5% | 48.4% | DeepSeek V4 Flash 0731 |
| Tax Agent Bench (Vals AI) | 28.1% | 24.9% | DeepSeek V4 Flash 0731 |
| Legal Research Bench (Vals AI) | 30.3% | 38.0% | Grok 4.5 |
| Excel Modeling Benchmark (Vals AI) | 57.0% | 52.9% | DeepSeek V4 Flash 0731 |
| 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 |
| Time Horizon Index (Vals AI) | Not measured | 6.3% | Only Grok 4.5 measured |
| APEX-Agents | Not measured | 56.2% | Only Grok 4.5 measured |
| PostTrainBench | Not measured | 23.4% | Only Grok 4.5 measured |
| Vending-Bench 2 | Not measured | $3,887 | Only Grok 4.5 measured |
| LMArena Search | Not measured | 1216 | Only Grok 4.5 measured |
| Neither model has a result on the other 17 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index26
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | Leader |
|---|---|---|---|
| GPQA Diamond | 89.9% | 92.9% | Grok 4.5 |
| LiveBench reasoning | Not measured | 87.2% | Only Grok 4.5 measured |
| MMMU Pro (Vals AI) | Not measured | 61.8% | Only Grok 4.5 measured |
| GPQA Diamond (Artificial Analysis run) | 90.8% | 93.1% | Grok 4.5 |
| Humanity's Last Exam (Artificial Analysis run) | 38.6% | 42.7% | Grok 4.5 |
| GPQA Diamond (Epoch AI run) | 91.0% | 93.4% | Grok 4.5 |
| ARC-AGI | 89.0% | 87.2% | DeepSeek V4 Flash 0731 |
| ARC-AGI-2 | 61.4% | 52.6% | DeepSeek V4 Flash 0731 |
| SimpleBench | 61.1% | 70.0% | Grok 4.5 |
| Chess puzzles | 32.0% | 35.0% | Grok 4.5 |
| Mystery game puzzles | 34.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LMArena Text: hard prompts | Not measured | 1462 | Only Grok 4.5 measured |
| LMArena Text: expert | Not measured | 1465 | Only Grok 4.5 measured |
| LMArena Text: hard prompts in English | Not measured | 1464 | Only Grok 4.5 measured |
| LMArena Vision: diagrams | Not measured | 1304 | Only Grok 4.5 measured |
| LMArena Vision: homework | Not measured | 1295 | Only Grok 4.5 measured |
| Kagi LLM Benchmark | Not measured | 83.5% | Only Grok 4.5 measured |
| Neither model has a result on the other 5 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index21
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 94.4% | 97.8% | Grok 4.5 |
| FrontierMath, tiers 1 to 3 | 57.5% | 57.2% | DeepSeek V4 Flash 0731 |
| FrontierMath, tier 4 | 24.4% | 24.4% | Tie |
| ProofBench | 56.0% | 31.0% | DeepSeek V4 Flash 0731 |
| LMArena Text: math | Not measured | 1460 | Only Grok 4.5 measured |
| LMArena Text: mathematical work | Not measured | 1482 | Only Grok 4.5 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | 89.2% | Grok 4.5 |
| LegalBench (Vals AI) | 77.7% | 86.0% | Grok 4.5 |
| CorpFin (Vals AI) | 61.9% | 67.4% | Grok 4.5 |
| TaxEval (Vals AI) | 70.7% | 71.7% | Grok 4.5 |
| MortgageTax (Vals AI) | Not measured | 61.8% | Only Grok 4.5 measured |
| MedCode (Vals AI) | 41.4% | 43.3% | Grok 4.5 |
| SimpleQA Verified | 33.6% | 48.3% | Grok 4.5 |
| LMArena Text: science | Not measured | 1464 | Only Grok 4.5 measured |
| LMArena Text: medicine and healthcare | Not measured | 1440 | Only Grok 4.5 measured |
| LMArena Text: legal and government | Not measured | 1463 | Only Grok 4.5 measured |
| LMArena Text: business and finance | Not measured | 1441 | Only Grok 4.5 measured |
| FACTS Benchmark Suite | Not measured | 50.2% | Only Grok 4.5 measured |
| Neither model has a result on the other 6 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | Leader |
|---|---|---|---|
| Design Arena: websites | 1238 | 1281 | Grok 4.5 |
| Design Arena: UI components | 1235 | 1279 | Grok 4.5 |
| Design Arena: game dev | 1208 | 1263 | Grok 4.5 |
| Design Arena: data viz | 1180 | 1266 | Grok 4.5 |
| Design Arena: 3D | 1209 | 1257 | Grok 4.5 |
| Design Arena: all code categories | 1230 | 1275 | Grok 4.5 |
| Design Arena: SVG | 1185 | 1215 | Grok 4.5 |
| Design Arena: ASCII art | 1090 | 1257 | Grok 4.5 |
| LiveBench design | Not measured | 78.1% | Only Grok 4.5 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index13
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 73.0% | Only Grok 4.5 measured |
| SAGE (Vals AI) | Not measured | 35.0% | Only Grok 4.5 measured |
| MedScribe (Vals AI) | 80.4% | 86.9% | Grok 4.5 |
| Vals Multimodal Index | Not measured | 63.4% | Only Grok 4.5 measured |
| Long context reasoning (AA-LCR) | 79.7% | 79.3% | DeepSeek V4 Flash 0731 |
| DTBench | 90.9% | 96.5% | Grok 4.5 |
| LMCA | 41.7% | 45.2% | Grok 4.5 |
| Surface Evolver Bench | Not measured | 74.4% | Only Grok 4.5 measured |
| Furniture assembly | Not measured | 22.5% | Only Grok 4.5 measured |
| LMArena Text | Not measured | 1448 | Only Grok 4.5 measured |
| LMArena Text: instruction following | Not measured | 1445 | Only Grok 4.5 measured |
| LMArena Text: creative writing | Not measured | 1442 | Only Grok 4.5 measured |
| LMArena Vision | Not measured | 1287 | Only Grok 4.5 measured |
| LMArena Text: English | Not measured | 1453 | Only Grok 4.5 measured |
| LMArena Text: non-English | Not measured | 1440 | Only Grok 4.5 measured |
| LMArena Text: Chinese | Not measured | 1495 | Only Grok 4.5 measured |
| LMArena Text: German | Not measured | 1448 | Only Grok 4.5 measured |
| LMArena Text: Japanese | Not measured | 1430 | Only Grok 4.5 measured |
| LMArena Text: Korean | Not measured | 1406 | Only Grok 4.5 measured |
| LMArena Text: French | Not measured | 1457 | Only Grok 4.5 measured |
| LMArena Text: Spanish | Not measured | 1451 | Only Grok 4.5 measured |
| LMArena Text: Russian | Not measured | 1449 | Only Grok 4.5 measured |
| LMArena Text: Polish | Not measured | 1432 | Only Grok 4.5 measured |
| LMArena Text: multi-turn | Not measured | 1456 | Only Grok 4.5 measured |
| LMArena Text: longer queries | Not measured | 1462 | Only Grok 4.5 measured |
| LMArena Text: writing and language | Not measured | 1445 | Only Grok 4.5 measured |
| LMArena Text: entertainment and media | Not measured | 1434 | Only Grok 4.5 measured |
| LMArena Vision: English | Not measured | 1285 | Only Grok 4.5 measured |
| LMArena Vision: Chinese | Not measured | 1345 | Only Grok 4.5 measured |
| LMArena Vision: reading text | Not measured | 1293 | Only Grok 4.5 measured |
| LMArena Vision: humor | Not measured | 1276 | Only Grok 4.5 measured |
| LMArena Vision: creative writing | Not measured | 1293 | Only Grok 4.5 measured |
| LMArena Document | Not measured | 1452 | Only Grok 4.5 measured |
| Neither model has a result on the other 13 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | Grok 4.5 | 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 0731 costs less
Repository review
50K input + 3K output tokens
DeepSeek V4 Flash 0731 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4 Flash 0731 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 | SpaceXAI Grok 4.5 |
|---|---|---|
| Maker | DeepSeek | xAI |
| Weights | Open weights | No public weights listed |
| Released | July 31, 2026 | July 8, 2026 |
| Context window | 1.05M | 500K |
| Input $/Mtok | $0.14 | $2.00 |
| Output $/Mtok | $0.28 | $6.00 |
| Coding index | 51.8 | 58.6 |
| Independent measurements behind it | 4 | 6 |
| Cost per solved task | $0.01 | $0.62 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Flash 0731 better than Grok 4.5 for coding?
Pick Grok 4.5 for coding: it scores 58.6 on the Vibe Coding Index against 51.8 for DeepSeek V4 Flash 0731. Pick DeepSeek V4 Flash 0731 for output tokens 21.4x cheaper ($0.28 vs $6.00 per million) and a lower cost per solved task ($0.01 against $0.62).
Which is cheaper, DeepSeek V4 Flash 0731 or Grok 4.5?
DeepSeek V4 Flash 0731 costs $0.14 per million input tokens and $0.28 per million output tokens; Grok 4.5 costs $2.00 and $6.00. Per solved SWE-bench Verified task, DeepSeek V4 Flash 0731 costs $0.01 and Grok 4.5 $0.62.
Which has the bigger context window, DeepSeek V4 Flash 0731 or Grok 4.5?
DeepSeek V4 Flash 0731, with 1.05M tokens against 500K.
How many benchmarks is the DeepSeek V4 Flash 0731 vs Grok 4.5 comparison based on?
52 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 Grok 4.5's on 6. Where only one model has a result, the page says so instead of naming a winner.