LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

R1 0528 vs Grok 4.6

Updated October 21, 2026. 46 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace R1 0528 Replace Grok 4.6
R1 0528DeepSeek
Coding
71.4
Overall
13.1
Output $/M
$2.15
4
Grok 4.6xAI
Coding
76.4
Overall
44.3
Output $/M
$6.00
42
vs boards won of 46 shared

Decision reading

Pick Grok 4.6 for coding: it scores 76.4 on the Vibe Coding Index against 71.4 for R1 0528.

Pick R1 0528 when output tokens cost 2.8x less, or when what matters is LMArena Text: medicine and healthcare or LMArena Text: English, where it leads.

Model AR1 0528DeepSeek71.4/1001 independent benchmarkNo cost per solved taskModel BGrok 4.6xAI76.4/1002 independent benchmarks$0.82 per solved task
Shared results
46
R1 0528 only
13
Grok 4.6 only
77
Like-for-like categories
7 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.

At a glance

Coding index
R1 0528 71.4 from 1 benchmark, Grok 4.6 76.4 from 2 benchmarks.
Widest category
Intelligence, 31.2 points apart on average over 1 shared benchmark.
Widest single gap
ARC-AGI: R1 0528 21.2% against Grok 4.6 87.5%, 66.3 points apart on a 0 to 100 scale.
Output price
R1 0528 $2.15 against Grok 4.6 $6.00 per million output tokens, 2.8x apart.
Context window
R1 0528 takes 164K tokens, Grok 4.6 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.

Boards won, where both have a result
CategoryR1 0528BarsGrok 4.6Ties
Coding04
Intelligence01
Reasoning09
Math03
Knowledge13
Web design07
General315
All shared boards442
CodingLike-for-likeFour of these boards feed the Vibe Coding Index04
Coding benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
Aider polyglot71.4%Not measuredOnly R1 0528 measured
LiveBench codingNot measured89.1%Only Grok 4.6 measured
LMArena WebDevNot measured1617Only Grok 4.6 measured
AA Coding IndexNot measured76.8Only Grok 4.6 measured
LiveCodeBench (Vals AI)Not measured88.2%Only Grok 4.6 measured
SWE-bench Verified (Vals AI)Not measured95.6%Only Grok 4.6 measured
Terminal-Bench 4 (Vals AI)Not measured17.2%Only Grok 4.6 measured
Vibe Code Bench (Vals AI)Not measured76.2%Only Grok 4.6 measured
Code migration (Vals AI)Not measured44.6%Only Grok 4.6 measured
ProgramBench (Vals AI)Not measured1.0%Only Grok 4.6 measured
IOI (Vals AI)Not measured47.6%Only Grok 4.6 measured
Terminal-Bench 2.1 (Vals AI)Not measured78.3%Only Grok 4.6 measured
Vibe Code Bench 1-100 (Vals AI)Not measured14.8%Only Grok 4.6 measured
SkillsBench (Vals AI)Not measured55.8%Only Grok 4.6 measured
CyberBench (Vals AI)Not measured67.7%Only Grok 4.6 measured
LiveCodeBench (Artificial Analysis run)77.0%Not measuredOnly R1 0528 measured
SciCodeNot measured56.5%Only Grok 4.6 measured
Terminal-Bench Hard (Artificial Analysis run)15.9%Not measuredOnly R1 0528 measured
FrontierCodeNot measured48.0%Only Grok 4.6 measured
FrontierSWENot measured25.3%Only Grok 4.6 measured
WeirdML41.6%67.3%Grok 4.6
ALE-Bench804.11508.1Grok 4.6
LMArena Text: coding14271466Grok 4.6
LMArena Text: software and IT14351457Grok 4.6
LMArena WebDev: ReactNot measured1631Only Grok 4.6 measured
LMArena WebDev: HTMLNot measured1597Only Grok 4.6 measured
LMArena WebDev: from an imageNot measured1603Only Grok 4.6 measured
Neither model has a result on the other 11 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
AA Intelligence Index13.144.3Grok 4.6
Epoch Capabilities IndexNot measured156.6Only Grok 4.6 measured
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index25 boards, none shared
Agentic work benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
AA Agentic IndexNot measured53.0Only Grok 4.6 measured
Design Arena: agents building web appsNot measured1247Only Grok 4.6 measured
Design Arena: agents building full-stack appsNot measured1264Only Grok 4.6 measured
Design Arena: agents building mobile appsNot measured1234Only Grok 4.6 measured
Design Arena: agents building Android appsNot measured1297Only Grok 4.6 measured
Design Arena: agents building Godot gamesNot measured1276Only Grok 4.6 measured
Design Arena: agents building gamesNot measured1208Only Grok 4.6 measured
Design Arena: HTML slidesNot measured1249Only Grok 4.6 measured
Design Arena: PowerPoint slidesNot measured1241Only Grok 4.6 measured
LiveBench researchNot measured84.5%Only Grok 4.6 measured
LiveBench browserNot measured64.0%Only Grok 4.6 measured
LiveBench long agenticNot measured40.7%Only Grok 4.6 measured
Finance Agent (Vals AI)Not measured53.7%Only Grok 4.6 measured
Tax Agent Bench (Vals AI)Not measured33.1%Only Grok 4.6 measured
Legal Research Bench (Vals AI)Not measured48.1%Only Grok 4.6 measured
Excel Modeling Benchmark (Vals AI)Not measured62.7%Only Grok 4.6 measured
BioMysteryBench (Vals AI)Not measured72.2%Only Grok 4.6 measured
Terminal-Bench-Science (Vals AI)Not measured11.4%Only Grok 4.6 measured
Vals Science IndexNot measured38.1%Only Grok 4.6 measured
Time Horizon Index (Vals AI)Not measured5.8%Only Grok 4.6 measured
τ²-Bench Telecom36.5%Not measuredOnly R1 0528 measured
APEX-AgentsNot measured65.3%Only Grok 4.6 measured
METR time horizons53.8%Not measuredOnly R1 0528 measured
DeepResearch Bench35.1%Not measuredOnly R1 0528 measured
Vending-Bench 2Not measured$9,047Only Grok 4.6 measured
Neither model has a result on the other 14 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index09
Reasoning benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
GPQA DiamondNot measured94.7%Only Grok 4.6 measured
LiveBench reasoningNot measured91.2%Only Grok 4.6 measured
MysteryMechanism (Vals AI)Not measured30.6%Only Grok 4.6 measured
GPQA Diamond (Artificial Analysis run)81.3%94.9%Grok 4.6
Humanity's Last Exam (Artificial Analysis run)15.8%44.1%Grok 4.6
GPQA Diamond (Epoch AI run)76.3%94.0%Grok 4.6
ARC-AGI21.2%87.5%Grok 4.6
ARC-AGI-21.1%67.1%Grok 4.6
SimpleBench40.8%75.9%Grok 4.6
Chess puzzlesNot measured33.0%Only Grok 4.6 measured
Mystery game puzzlesNot measured34.0%Only Grok 4.6 measured
EBR-benchNot measured30.5%Only Grok 4.6 measured
LMArena Text: hard prompts14151448Grok 4.6
LMArena Text: expert13941469Grok 4.6
LMArena Text: hard prompts in English14221445Grok 4.6
LMArena Vision: diagramsNot measured1285Only Grok 4.6 measured
LMArena Vision: homeworkNot measured1283Only Grok 4.6 measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index03
Math benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
AA Math Index76.0Not measuredOnly R1 0528 measured
AIME 2025 (Artificial Analysis run)76.0%Not measuredOnly R1 0528 measured
OTIS Mock AIME66.4%99.2%Grok 4.6
FrontierMath, tiers 1 to 3Not measured66.0%Only Grok 4.6 measured
FrontierMath, tier 4Not measured31.7%Only Grok 4.6 measured
ProofBenchNot measured51.0%Only Grok 4.6 measured
MATH level 596.6%Not measuredOnly R1 0528 measured
LMArena Text: math14001417Grok 4.6
LMArena Text: mathematical work14151464Grok 4.6
Neither model has a result on the other 4 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index13
Knowledge benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
MMLU ProNot measured89.4%Only Grok 4.6 measured
LegalBench (Vals AI)Not measured86.3%Only Grok 4.6 measured
CorpFin (Vals AI)Not measured66.2%Only Grok 4.6 measured
TaxEval (Vals AI)Not measured71.1%Only Grok 4.6 measured
MortgageTax (Vals AI)Not measured64.2%Only Grok 4.6 measured
MedCode (Vals AI)Not measured44.7%Only Grok 4.6 measured
MMLU Pro (Artificial Analysis run)84.9%Not measuredOnly R1 0528 measured
SimpleQA VerifiedNot measured49.3%Only Grok 4.6 measured
LMArena Text: science14301437Grok 4.6
LMArena Text: medicine and healthcare14451420R1 0528
LMArena Text: legal and government14191429Grok 4.6
LMArena Text: business and finance14021416Grok 4.6
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index07
Web design benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
Design Arena: websites11491289Grok 4.6
Design Arena: UI components11021294Grok 4.6
Design Arena: game dev10981296Grok 4.6
Design Arena: data viz11821283Grok 4.6
Design Arena: 3D11111277Grok 4.6
Design Arena: all code categories11391292Grok 4.6
Design Arena: SVG10381239Grok 4.6
Design Arena: ASCII artNot measured1278Only Grok 4.6 measured
LiveBench designNot measured79.8%Only Grok 4.6 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index315
General benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
LiveBench data analysisNot measured75.3%Only Grok 4.6 measured
Public Benefits Bench (Vals AI)Not measured66.8%Only Grok 4.6 measured
SAGE (Vals AI)Not measured28.9%Only Grok 4.6 measured
MedScribe (Vals AI)Not measured86.5%Only Grok 4.6 measured
IFBench39.6%Not measuredOnly R1 0528 measured
Long context reasoning (AA-LCR)55.7%81.0%Grok 4.6
DTBenchNot measured97.3%Only Grok 4.6 measured
LMCANot measured48.5%Only Grok 4.6 measured
Fiction.LiveBench75.0%Not measuredOnly R1 0528 measured
Creative writing (Lech Mazur)81.9%Not measuredOnly R1 0528 measured
Furniture assemblyNot measured40.0%Only Grok 4.6 measured
LMArena Text14281429Grok 4.6
LMArena Text: instruction following13821431Grok 4.6
LMArena Text: creative writing14051426Grok 4.6
LMArena VisionNot measured1263Only Grok 4.6 measured
LMArena Text: English14371433R1 0528
LMArena Text: non-English14121420Grok 4.6
LMArena Text: Chinese14421476Grok 4.6
LMArena Text: German14041426Grok 4.6
LMArena Text: Japanese13911378R1 0528
LMArena Text: Korean13601391Grok 4.6
LMArena Text: French14171455Grok 4.6
LMArena Text: Spanish14111414Grok 4.6
LMArena Text: Russian14231421R1 0528
LMArena Text: Polish14281430Grok 4.6
LMArena Text: multi-turn14051425Grok 4.6
LMArena Text: longer queries13911456Grok 4.6
LMArena Text: writing and language14021429Grok 4.6
LMArena Text: entertainment and media14001421Grok 4.6
LMArena Vision: EnglishNot measured1270Only Grok 4.6 measured
LMArena Vision: ChineseNot measured1289Only Grok 4.6 measured
LMArena Vision: reading textNot measured1272Only Grok 4.6 measured
LMArena Vision: humorNot measured1270Only Grok 4.6 measured
LMArena Vision: creative writingNot measured1262Only Grok 4.6 measured
LMArena DocumentNot measured1452Only Grok 4.6 measured
Neither model has a result on the other 11 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for R1 0528 and Grok 4.6
BenchmarkR1 0528Grok 4.6Leader
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

R1 0528$0.0016Fits in one request
Grok 4.6$0.0050Fits in one request

R1 0528 costs less

Repository review

50K input + 3K output tokens

R1 0528$0.031Fits in one request
Grok 4.6$0.118Fits in one request

R1 0528 costs less

Agent loop

200K input + 10K output tokens

R1 0528$0.121Exceeds the context window
Grok 4.6$0.460Fits in one request

R1 0528 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.

Specifications of R1 0528 and Grok 4.6
SpecificationDeepSeek R1 0528SpaceXAI Grok 4.6
MakerDeepSeekxAI
WeightsOpen weightsNo public weights listed
ReleasedMay 28, 2025August 12, 2026
Context window164K500K
Input $/Mtok$0.50$2.00
Output $/Mtok$2.15$6.00
Coding index71.476.4
Independent measurements behind it12
Cost per solved taskNot computable$0.82
SourceOpenRouter listingOpenRouter listing

Questions

Is R1 0528 better than Grok 4.6 for coding?

Pick Grok 4.6 for coding: it scores 76.4 on the Vibe Coding Index against 71.4 for R1 0528. Pick R1 0528 when output tokens cost 2.8x less, or when what matters is LMArena Text: medicine and healthcare or LMArena Text: English, where it leads.

Which is cheaper, R1 0528 or Grok 4.6?

R1 0528 costs $0.50 per million input tokens and $2.15 per million output tokens; Grok 4.6 costs $2.00 and $6.00. Cost per solved task is known for Grok 4.6 only.

Which has the bigger context window, R1 0528 or Grok 4.6?

Grok 4.6, with 500K tokens against 164K.

How many benchmarks is the R1 0528 vs Grok 4.6 comparison based on?

46 benchmarks have a result for both models as of October 21, 2026. R1 0528's coding index rests on 1 independent benchmark and Grok 4.6's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next