Grok 4.7 vs Step 5 Preview
Updated October 21, 2026. 61 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick Grok 4.7 for coding: it scores 75.0 on the Vibe Coding Index against 71.4 for Step 5 Preview.
Pick Step 5 Preview when output tokens cost 2.2x less, or when the context window is larger, 1.05M against 500K tokens.
- Shared results
- 61
- Grok 4.7 only
- 38
- Step 5 Preview only
- 0
- Like-for-like categories
- 8 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
Grok 4.7
Grok 4.7 scores 75.0 against 71.4.
2 benchmarks behind each score
Intelligence
AA Intelligence Index
Grok 4.7
Grok 4.7 leads, 46.4 against 43.7.
Same board, both measured
Web design
Design Arena: websites
No call
Only Grok 4.7 has a Design Arena: websites result, 1255.
Directional only
General
Vals Index
No call
Only Grok 4.7 has a Vals Index result, 58.6%.
Directional only
Long documents
Listed context window
Step 5 Preview
Step 5 Preview takes 1.05M tokens against 500K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Step 5 Preview
Grok 4.7 $0.0050, Step 5 Preview $0.0024 per turn.
List prices
At a glance
- Coding index
- Grok 4.7 75.0 from 2 benchmarks, Step 5 Preview 71.4 from 2 benchmarks.
- Widest category
- Agentic work, 10.5 points apart on average over 11 shared benchmarks.
- Widest single gap
- LiveBench browser: Grok 4.7 60.0% against Step 5 Preview 24.0%, 36.0 points apart on a 0 to 100 scale.
- Output price
- Grok 4.7 $6.00 against Step 5 Preview $2.70 per million output tokens, 2.2x apart.
- Context window
- Grok 4.7 takes 500K tokens, Step 5 Preview 1.05M.
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 | Grok 4.7 | Bars | Step 5 Preview | Ties |
|---|---|---|---|---|
| Coding | 8 | 5 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 11 | 0 | ||
| Reasoning | 2 | 6 | ||
| Math | 0 | 3 | ||
| Knowledge | 0 | 4 | ||
| Web design | 1 | 0 | ||
| General | 2 | 18 | ||
| All shared boards | 25 | 36 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index85
| Benchmark | Grok 4.7 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench coding | 84.5% | 85.6% | Step 5 Preview |
| LMArena WebDev | 1637 | 1567 | Grok 4.7 |
| Terminal-Bench 4 (Vals AI) | 28.8% | 31.8% | Step 5 Preview |
| Vibe Code Bench (Vals AI) | 86.2% | 70.4% | Grok 4.7 |
| Code migration (Vals AI) | 44.8% | 39.0% | Grok 4.7 |
| ProgramBench (Vals AI) | 0.5% | Not measured | Only Grok 4.7 measured |
| IOI (Vals AI) | 57.7% | 38.1% | Grok 4.7 |
| Terminal-Bench 2.1 (Vals AI) | 73.4% | Not measured | Only Grok 4.7 measured |
| SWE-bench Multimodal (Vals AI) | 42.9% | 40.8% | Grok 4.7 |
| CyberBench (Vals AI) | 69.5% | Not measured | Only Grok 4.7 measured |
| SciCode | 57.8% | 58.9% | Step 5 Preview |
| FrontierCode | 47.6% | Not measured | Only Grok 4.7 measured |
| FrontierSWE | 29.5% | Not measured | Only Grok 4.7 measured |
| LMArena Text: coding | 1432 | 1485 | Step 5 Preview |
| LMArena Text: software and IT | 1421 | 1473 | Step 5 Preview |
| LMArena WebDev: React | 1663 | 1583 | Grok 4.7 |
| LMArena WebDev: HTML | 1627 | 1556 | Grok 4.7 |
| LMArena WebDev: from an image | 1586 | 1550 | Grok 4.7 |
| Neither model has a result on the other 20 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Grok 4.7 | Step 5 Preview | Leader |
|---|---|---|---|
| AA Intelligence Index | 46.4 | 43.7 | Grok 4.7 |
| Epoch Capabilities Index | 154.4 | Not measured | Only Grok 4.7 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index110
| Benchmark | Grok 4.7 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench research | 86.4% | 72.8% | Grok 4.7 |
| LiveBench browser | 60.0% | 24.0% | Grok 4.7 |
| LiveBench long agentic | 50.1% | 27.5% | Grok 4.7 |
| SREBench (Vals AI) | 3.8% | 0.4% | Grok 4.7 |
| Finance Agent (Vals AI) | 52.3% | 50.7% | Grok 4.7 |
| Tax Agent Bench (Vals AI) | 34.2% | 27.3% | Grok 4.7 |
| Legal Research Bench (Vals AI) | 47.1% | 40.4% | Grok 4.7 |
| Harvey's Legal Agent Benchmark (Vals AI) | 18.3% | Not measured | Only Grok 4.7 measured |
| Excel Modeling Benchmark (Vals AI) | 67.0% | 57.1% | Grok 4.7 |
| BioMysteryBench (Vals AI) | 69.3% | 67.4% | Grok 4.7 |
| Terminal-Bench-Science (Vals AI) | 10.0% | 2.9% | Grok 4.7 |
| Vals Science Index | 34.8% | 29.6% | Grok 4.7 |
| Robo-bench (Vals AI) | 0.2% | Not measured | Only Grok 4.7 measured |
| APEX-Agents | 54.6% | Not measured | Only Grok 4.7 measured |
| Vending-Bench 2 | $10,537 | Not measured | Only Grok 4.7 measured |
| Neither model has a result on the other 24 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index26
| Benchmark | Grok 4.7 | Step 5 Preview | Leader |
|---|---|---|---|
| LiveBench reasoning | 82.7% | 83.8% | Step 5 Preview |
| MysteryMechanism (Vals AI) | 25.2% | 18.5% | Grok 4.7 |
| Humanity's Last Exam (Artificial Analysis run) | 43.1% | 46.5% | Step 5 Preview |
| GPQA Diamond (Epoch AI run) | 92.7% | Not measured | Only Grok 4.7 measured |
| ARC-AGI | 90.2% | Not measured | Only Grok 4.7 measured |
| ARC-AGI-2 | 61.4% | Not measured | Only Grok 4.7 measured |
| Chess puzzles | 41.0% | Not measured | Only Grok 4.7 measured |
| Mystery game puzzles | 29.0% | Not measured | Only Grok 4.7 measured |
| LMArena Text: hard prompts | 1415 | 1461 | Step 5 Preview |
| LMArena Text: expert | 1439 | 1485 | Step 5 Preview |
| LMArena Text: hard prompts in English | 1427 | 1454 | Step 5 Preview |
| LMArena Vision: diagrams | 1254 | 1308 | Step 5 Preview |
| LMArena Vision: homework | 1309 | 1265 | Grok 4.7 |
| Neither model has a result on the other 8 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | Grok 4.7 | Step 5 Preview | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.1% | Not measured | Only Grok 4.7 measured |
| FrontierMath, tiers 1 to 3 | 53.0% | Not measured | Only Grok 4.7 measured |
| FrontierMath, tier 4 | 17.1% | Not measured | Only Grok 4.7 measured |
| ProofBench | 26.0% | 42.0% | Step 5 Preview |
| LMArena Text: math | 1414 | 1458 | Step 5 Preview |
| LMArena Text: mathematical work | 1435 | 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 index04
| Benchmark | Grok 4.7 | Step 5 Preview | Leader |
|---|---|---|---|
| LegalBench (Vals AI) | 84.4% | Not measured | Only Grok 4.7 measured |
| MedCode (Vals AI) | 49.5% | Not measured | Only Grok 4.7 measured |
| SimpleQA Verified | 56.0% | Not measured | Only Grok 4.7 measured |
| LMArena Text: science | 1419 | 1454 | Step 5 Preview |
| LMArena Text: medicine and healthcare | 1395 | 1439 | Step 5 Preview |
| LMArena Text: legal and government | 1423 | 1437 | Step 5 Preview |
| LMArena Text: business and finance | 1389 | 1435 | Step 5 Preview |
| Neither model has a result on the other 10 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Grok 4.7 | Step 5 Preview | Leader |
|---|---|---|---|
| Design Arena: websites | 1255 | Not measured | Only Grok 4.7 measured |
| Design Arena: UI components | 1271 | Not measured | Only Grok 4.7 measured |
| Design Arena: game dev | 1314 | Not measured | Only Grok 4.7 measured |
| Design Arena: data viz | 1227 | Not measured | Only Grok 4.7 measured |
| Design Arena: 3D | 1269 | Not measured | Only Grok 4.7 measured |
| Design Arena: all code categories | 1271 | Not measured | Only Grok 4.7 measured |
| LiveBench design | 78.0% | 76.9% | Grok 4.7 |
| Neither model has a result on the other 2 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index218
| Benchmark | Grok 4.7 | Step 5 Preview | Leader |
|---|---|---|---|
| Vals Index | 58.6% | Not measured | Only Grok 4.7 measured |
| LiveBench data analysis | 76.9% | 63.5% | Grok 4.7 |
| Public Benefits Bench (Vals AI) | 65.6% | 58.1% | Grok 4.7 |
| SAGE (Vals AI) | 40.8% | Not measured | Only Grok 4.7 measured |
| MedScribe (Vals AI) | 89.4% | Not measured | Only Grok 4.7 measured |
| Long context reasoning (AA-LCR) | 80.0% | 88.3% | Step 5 Preview |
| DTBench | 96.0% | Not measured | Only Grok 4.7 measured |
| LMCA | 49.4% | Not measured | Only Grok 4.7 measured |
| Furniture assembly | 20.8% | Not measured | Only Grok 4.7 measured |
| LMArena Text | 1399 | 1440 | Step 5 Preview |
| LMArena Text: instruction following | 1406 | 1438 | Step 5 Preview |
| LMArena Text: creative writing | 1381 | 1409 | Step 5 Preview |
| LMArena Vision | 1231 | 1265 | Step 5 Preview |
| LMArena Text: English | 1411 | 1447 | Step 5 Preview |
| LMArena Text: non-English | 1386 | 1428 | Step 5 Preview |
| LMArena Text: Chinese | 1449 | 1500 | Step 5 Preview |
| LMArena Text: German | 1366 | Not measured | Only Grok 4.7 measured |
| LMArena Text: Japanese | 1359 | 1383 | Step 5 Preview |
| LMArena Text: Korean | 1383 | Not measured | Only Grok 4.7 measured |
| LMArena Text: French | 1431 | 1438 | Step 5 Preview |
| LMArena Text: Spanish | 1406 | 1441 | Step 5 Preview |
| LMArena Text: Russian | 1399 | 1433 | Step 5 Preview |
| LMArena Text: Polish | 1420 | Not measured | Only Grok 4.7 measured |
| LMArena Text: multi-turn | 1394 | 1437 | Step 5 Preview |
| LMArena Text: longer queries | 1416 | 1449 | Step 5 Preview |
| LMArena Text: writing and language | 1394 | 1425 | Step 5 Preview |
| LMArena Text: entertainment and media | 1376 | 1412 | Step 5 Preview |
| LMArena Vision: English | 1229 | 1290 | Step 5 Preview |
| LMArena Vision: Chinese | 1250 | Not measured | Only Grok 4.7 measured |
| LMArena Vision: reading text | 1241 | 1282 | Step 5 Preview |
| LMArena Vision: creative writing | 1217 | Not measured | Only Grok 4.7 measured |
| Neither model has a result on the other 15 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Grok 4.7 | 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 | SpaceXAI Grok 4.7 | StepFun Step 5 Preview |
|---|---|---|
| Maker | xAI | StepFun |
| Weights | No public weights listed | Open weights |
| Released | September 21, 2026 | October 8, 2026 |
| Context window | 500K | 1.05M |
| Input $/Mtok | $2.00 | $1.00 |
| Output $/Mtok | $6.00 | $2.70 |
| Coding index | 75.0 | 71.4 |
| Independent measurements behind it | 2 | 2 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Grok 4.7 better than Step 5 Preview for coding?
Pick Grok 4.7 for coding: it scores 75.0 on the Vibe Coding Index against 71.4 for Step 5 Preview. Pick Step 5 Preview when output tokens cost 2.2x less, or when the context window is larger, 1.05M against 500K tokens.
Which is cheaper, Grok 4.7 or Step 5 Preview?
Grok 4.7 costs $2.00 per million input tokens and $6.00 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, Grok 4.7 or Step 5 Preview?
Step 5 Preview, with 1.05M tokens against 500K.
How many benchmarks is the Grok 4.7 vs Step 5 Preview comparison based on?
61 benchmarks have a result for both models as of October 21, 2026. Grok 4.7's coding index rests on 2 independent benchmarks and Step 5 Preview's on 2. Where only one model has a result, the page says so instead of naming a winner.