Claude Opus 4.6 vs Grok 4.20
Claude Opus 4.6 leads Grok 4.20 on coding, 57.2 to 33.3 on the Vibe Coding Index, and Grok 4.20 costs 10.0x less per output token ($2.50 vs $25.00 per million). Updated October 21, 2026; 46 benchmarks have a result for both.
Claude Opus 4.6 or Grok 4.20: which to pick
Pick Claude Opus 4.6 for coding: it scores 57.2 on the Vibe Coding Index against 33.3 for Grok 4.20.
Pick Grok 4.20 for output tokens 10.0x cheaper ($2.50 vs $25.00 per million) and a lower cost per solved task ($0.72 against $1.56).
- Shared results
- 46
- Claude Opus 4.6 only
- 64
- Grok 4.20 only
- 9
- 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
Claude Opus 4.6
Claude Opus 4.6 scores 57.2 against 33.3.
4 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
Claude Opus 4.6
Claude Opus 4.6 leads, 31.9 against 25.7.
Same board, both measured
Reasoning
GPQA Diamond
Claude Opus 4.6
Claude Opus 4.6 leads, 89.7% against 88.6%.
Same board, both measured
Math
AIME
Grok 4.20
Grok 4.20 leads, 96.5% against 95.6%.
Same board, both measured
Knowledge
MMLU Pro
Claude Opus 4.6
Claude Opus 4.6 leads, 89.1% against 86.3%.
Same board, both measured
Web design
Design Arena: websites
Claude Opus 4.6
Claude Opus 4.6 leads, 1291 against 1229.
Same board, both measured
Long documents
Listed context window
Grok 4.20
Grok 4.20 takes 2M tokens against 1M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Grok 4.20
Claude Opus 4.6 $0.018, Grok 4.20 $0.0025 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Grok 4.20
Claude Opus 4.6 $1.56, Grok 4.20 $0.72.
Published attempt cost and pass rate
At a glance
- Coding index
- Claude Opus 4.6 57.2 from 4 benchmarks, Grok 4.20 33.3 from 2 benchmarks.
- Widest category
- Coding, 19.8 points apart on average over 6 shared benchmarks.
- Widest single gap
- Vibe Code Bench (Vals AI): Claude Opus 4.6 57.6% against Grok 4.20 4.1%, 53.5 points apart on a 0 to 100 scale.
- Output price
- Claude Opus 4.6 $25.00 against Grok 4.20 $2.50 per million output tokens, 10.0x apart.
- Context window
- Claude Opus 4.6 takes 1M tokens, Grok 4.20 2M.
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 | Claude Opus 4.6 | Bars | Grok 4.20 | Ties |
|---|---|---|---|---|
| Coding | 6 | 0 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 5 | 1 | ||
| Reasoning | 6 | 1 | ||
| Math | 0 | 1 | ||
| Knowledge | 8 | 0 | ||
| Web design | 8 | 0 | ||
| General | 6 | 2 | ||
| All shared boards | 41 | 5 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index60
| Benchmark | Claude Opus 4.6 | Grok 4.20 | Leader |
|---|---|---|---|
| SWE-bench Verified | 75.6% | Not measured | Only Claude Opus 4.6 measured |
| SWE-bench Multilingual | 72.0% | Not measured | Only Claude Opus 4.6 measured |
| LMArena WebDev | 1547 | Not measured | Only Claude Opus 4.6 measured |
| LiveCodeBench (Vals AI) | 84.7% | 84.3% | Claude Opus 4.6 |
| SWE-bench Verified (Vals AI) | 78.2% | 72.2% | Claude Opus 4.6 |
| Vibe Code Bench (Vals AI) | 57.6% | 4.1% | Claude Opus 4.6 |
| Code migration (Vals AI) | Not measured | 0.3% | Only Grok 4.20 measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 44.2% | Only Grok 4.20 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 48.5% | 37.9% | Claude Opus 4.6 |
| Terminal-Bench | 79.8% | 57.3% | Claude Opus 4.6 |
| FrontierCode | 26.6% | Not measured | Only Claude Opus 4.6 measured |
| GSO-Bench | 41.2% | Not measured | Only Claude Opus 4.6 measured |
| WeirdML | 78.0% | 52.3% | Claude Opus 4.6 |
| Cybench | 93.0% | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: coding | 1534 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: software and IT | 1527 | Not measured | Only Claude Opus 4.6 measured |
| LMArena WebDev: React | 1539 | Not measured | Only Claude Opus 4.6 measured |
| LMArena WebDev: HTML | 1559 | Not measured | Only Claude Opus 4.6 measured |
| LMArena WebDev: from an image | 1543 | Not measured | Only Claude Opus 4.6 measured |
| Neither model has a result on the other 20 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Claude Opus 4.6 | Grok 4.20 | Leader |
|---|---|---|---|
| AA Intelligence Index | 31.9 | 25.7 | Claude Opus 4.6 |
| Epoch Capabilities Index | 155.7 | 152.1 | Claude Opus 4.6 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index51
| Benchmark | Claude Opus 4.6 | Grok 4.20 | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1206 | 1144 | Claude Opus 4.6 |
| Design Arena: agents building full-stack apps | 1220 | 1054 | Claude Opus 4.6 |
| Design Arena: agents building mobile apps | 1207 | 1090 | Claude Opus 4.6 |
| Design Arena: agents building Android apps | 1196 | 1052 | Claude Opus 4.6 |
| Design Arena: agents building Godot games | Not measured | 1089 | Only Grok 4.20 measured |
| Design Arena: agents building games | 1212 | Not measured | Only Claude Opus 4.6 measured |
| Design Arena: HTML slides | Not measured | 1180 | Only Grok 4.20 measured |
| Finance Agent (Vals AI) | Not measured | 28.5% | Only Grok 4.20 measured |
| Tax Agent Bench (Vals AI) | Not measured | 7.0% | Only Grok 4.20 measured |
| Legal Research Bench (Vals AI) | Not measured | 13.9% | Only Grok 4.20 measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 11.7% | Only Grok 4.20 measured |
| τ²-Bench Telecom | 92.1% | 93.0% | Grok 4.20 |
| APEX-Agents | 46.3% | Not measured | Only Claude Opus 4.6 measured |
| METR time horizons | 78.9% | Not measured | Only Claude Opus 4.6 measured |
| Remote Labor Index | 4.2% | Not measured | Only Claude Opus 4.6 measured |
| DeepResearch Bench | 55.3% | Not measured | Only Claude Opus 4.6 measured |
| Vending-Bench 2 | $8,018 | $4,663 | Claude Opus 4.6 |
| Neither model has a result on the other 24 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index61
| Benchmark | Claude Opus 4.6 | Grok 4.20 | Leader |
|---|---|---|---|
| GPQA Diamond | 89.7% | 88.6% | Claude Opus 4.6 |
| MMMU Pro (Vals AI) | 83.9% | 83.5% | Claude Opus 4.6 |
| GPQA Diamond (Artificial Analysis run) | 89.6% | 91.1% | Grok 4.20 |
| Humanity's Last Exam (Artificial Analysis run) | 39.9% | 34.5% | Claude Opus 4.6 |
| GPQA Diamond (Epoch AI run) | 90.5% | Not measured | Only Claude Opus 4.6 measured |
| ARC-AGI | 94.0% | 89.5% | Claude Opus 4.6 |
| ARC-AGI-2 | 69.2% | 65.1% | Claude Opus 4.6 |
| SimpleBench | 67.6% | Not measured | Only Claude Opus 4.6 measured |
| Chess puzzles | 21.0% | Not measured | Only Claude Opus 4.6 measured |
| Mystery game puzzles | 25.0% | Not measured | Only Claude Opus 4.6 measured |
| EBR-bench | 12.7% | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: hard prompts | 1525 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: expert | 1544 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: hard prompts in English | 1530 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision: diagrams | 1340 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision: homework | 1337 | Not measured | Only Claude Opus 4.6 measured |
| Kagi LLM Benchmark | 83.6% | 75.0% | Claude Opus 4.6 |
| Neither model has a result on the other 5 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Claude Opus 4.6 | Grok 4.20 | Leader |
|---|---|---|---|
| AIME | 95.6% | 96.5% | Grok 4.20 |
| OTIS Mock AIME | 94.4% | Not measured | Only Claude Opus 4.6 measured |
| FrontierMath, tiers 1 to 3 | 66.0% | Not measured | Only Claude Opus 4.6 measured |
| FrontierMath, tier 4 | 26.8% | Not measured | Only Claude Opus 4.6 measured |
| ProofBench | 50.0% | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: math | 1518 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: mathematical work | 1525 | Not measured | Only Claude Opus 4.6 measured |
| Neither model has a result on the other 6 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index80
| Benchmark | Claude Opus 4.6 | Grok 4.20 | Leader |
|---|---|---|---|
| MMLU Pro | 89.1% | 86.3% | Claude Opus 4.6 |
| LegalBench (Vals AI) | 85.3% | 77.7% | Claude Opus 4.6 |
| CaseLaw (Vals AI) | 62.1% | 54.4% | Claude Opus 4.6 |
| CorpFin (Vals AI) | 67.0% | 63.7% | Claude Opus 4.6 |
| TaxEval (Vals AI) | 76.0% | 74.1% | Claude Opus 4.6 |
| MortgageTax (Vals AI) | 68.5% | 45.4% | Claude Opus 4.6 |
| MedQA (Vals AI) | 95.4% | 94.5% | Claude Opus 4.6 |
| MedCode (Vals AI) | 49.1% | 32.2% | Claude Opus 4.6 |
| SimpleQA Verified | 47.0% | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: science | 1519 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: medicine and healthcare | 1504 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: legal and government | 1512 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: business and finance | 1497 | Not measured | Only Claude Opus 4.6 measured |
| FACTS Benchmark Suite | 41.5% | Not measured | Only Claude Opus 4.6 measured |
| Neither model has a result on the other 4 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index80
| Benchmark | Claude Opus 4.6 | Grok 4.20 | Leader |
|---|---|---|---|
| Design Arena: websites | 1291 | 1229 | Claude Opus 4.6 |
| Design Arena: UI components | 1284 | 1199 | Claude Opus 4.6 |
| Design Arena: game dev | 1278 | 1192 | Claude Opus 4.6 |
| Design Arena: data viz | 1279 | 1217 | Claude Opus 4.6 |
| Design Arena: 3D | 1274 | 1193 | Claude Opus 4.6 |
| Design Arena: all code categories | 1289 | 1221 | Claude Opus 4.6 |
| Design Arena: SVG | 1228 | 1164 | Claude Opus 4.6 |
| Design Arena: ASCII art | 1255 | 1174 | Claude Opus 4.6 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index62
| Benchmark | Claude Opus 4.6 | Grok 4.20 | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 51.6% | 38.2% | Claude Opus 4.6 |
| MedScribe (Vals AI) | 86.7% | 63.4% | Claude Opus 4.6 |
| Vals Multimodal Index | Not measured | 39.1% | Only Grok 4.20 measured |
| IFBench | 53.1% | 81.2% | Grok 4.20 |
| Long context reasoning (AA-LCR) | 78.0% | 69.0% | Claude Opus 4.6 |
| DTBench | 91.2% | 90.1% | Claude Opus 4.6 |
| LMCA | 55.8% | 38.7% | Claude Opus 4.6 |
| CL-bench | 20.7% | 22.2% | Grok 4.20 |
| Furniture assembly | 28.3% | Not measured | Only Claude Opus 4.6 measured |
| CL-bench Life | 17.0% | 11.9% | Claude Opus 4.6 |
| LMArena Text | 1501 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: instruction following | 1521 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: creative writing | 1505 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision | 1316 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: English | 1511 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: non-English | 1488 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: Chinese | 1551 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: German | 1500 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: Japanese | 1488 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: Korean | 1464 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: French | 1512 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: Spanish | 1510 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: Russian | 1497 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: Polish | 1499 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: multi-turn | 1511 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: longer queries | 1518 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: writing and language | 1502 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Text: entertainment and media | 1492 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision: English | 1324 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision: Chinese | 1370 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision: reading text | 1326 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision: humor | 1314 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision: recognizing things | 1274 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Vision: creative writing | 1331 | Not measured | Only Claude Opus 4.6 measured |
| LMArena Document | 1507 | Not measured | Only Claude Opus 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
| Benchmark | Claude Opus 4.6 | Grok 4.20 | 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
Grok 4.20 costs less
Repository review
50K input + 3K output tokens
Grok 4.20 costs less
Agent loop
200K input + 10K output tokens
Grok 4.20 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 | Anthropic Claude Opus 4.6 | SpaceXAI Grok 4.20 |
|---|---|---|
| Maker | Anthropic | xAI |
| Weights | No public weights listed | No public weights listed |
| Released | February 4, 2026 | March 31, 2026 |
| Context window | 1M | 2M |
| Input $/Mtok | $5.00 | $1.25 |
| Output $/Mtok | $25.00 | $2.50 |
| Coding index | 57.2 | 33.3 |
| Independent measurements behind it | 4 | 2 |
| Cost per solved task | $1.56 | $0.72 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Opus 4.6 better than Grok 4.20 for coding?
Pick Claude Opus 4.6 for coding: it scores 57.2 on the Vibe Coding Index against 33.3 for Grok 4.20. Pick Grok 4.20 for output tokens 10.0x cheaper ($2.50 vs $25.00 per million) and a lower cost per solved task ($0.72 against $1.56).
Which is cheaper, Claude Opus 4.6 or Grok 4.20?
Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; Grok 4.20 costs $1.25 and $2.50. Per solved SWE-bench Verified task, Claude Opus 4.6 costs $1.56 and Grok 4.20 $0.72.
Which has the bigger context window, Claude Opus 4.6 or Grok 4.20?
Grok 4.20, with 2M tokens against 1M.
How many benchmarks is the Claude Opus 4.6 vs Grok 4.20 comparison based on?
46 benchmarks have a result for both models as of October 21, 2026. Claude Opus 4.6's coding index rests on 4 independent benchmarks and Grok 4.20's on 2. Where only one model has a result, the page says so instead of naming a winner.