Skip to main content
BenchLM
Advertise

Model comparison

Claude Opus 4.6 vs Grok 4.20

Data verified

Head-to-head evidence from 15 shared benchmark results across 4 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

68.59/100
Margin
13.9pts
← winning
54.68/100
3 category wins0 category wins

Public leaderboard positions: Claude Opus 4.6 #16 (Supported); Grok 4.20 #88 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude Opus 4.6 and Grok 4.20 share 15 comparable benchmark results. 3 of 8 categories are comparable. 31 results are unique to Claude Opus 4.6; 3 to Grok 4.20.

Updated July 23, 2026
Shared results
15
Claude Opus 4.6 only
31
Grok 4.20 only
3
Comparable categories
3 / 8

Pick Claude Opus 4.6 if you want the stronger benchmark profile. Grok 4.20 only becomes the better choice if you want the cheaper token bill or you need the larger 2M context window.

Confidence note. This is a partial-evidence comparison with 15 shared benchmark results across 4 evidence categories; 3 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Claude Opus 4.6 is clearly ahead on the BenchAlign aggregate, 68.59 to 54.68. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Claude Opus 4.6's sharpest advantage is in agentic, where it averages 73 against 47.1. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 65.4% to 47.1%.

Claude Opus 4.6 is also the more expensive model on tokens at $5.00 input / $25.00 output per 1M tokens, versus $2.00 input / $6.00 output per 1M tokens for Grok 4.20. That is roughly 4.2x on output cost alone. Grok 4.20 is the reasoning model in the pair, while Claude Opus 4.6 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Grok 4.20 gives you the larger context window at 2M, compared with 1M for Claude Opus 4.6.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Claude Opus 4.6 and Grok 4.20
CategoryClaude Opus 4.6ΔGrok 4.20
AgenticClaude Opus 4.673.0Margin← 25.9Grok 4.2047.1
MultimodalClaude Opus 4.677.3Margin← 7.2Grok 4.2070.1
CodingClaude Opus 4.668.1Margin← 1.0Grok 4.2067.1
ReasoningClaude Opus 4.6Not measuredMarginNo overlapGrok 4.2053.3
KnowledgeClaude Opus 4.669.1MarginNo overlapGrok 4.20Not measured
MathClaude Opus 4.636.3MarginNo overlapGrok 4.20Not measured

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · Claude Opus 4.6B · Grok 4.20

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 65.4%B 47.1%
    Winner: Claude Opus 4.6Δ 18.3
    Terminal-Bench 2.0: Claude Opus 4.6 scored 65.4%; Grok 4.20 scored 47.1%. Claude Opus 4.6 wins this benchmark.

  2. SWE-bench Verified

    Coding
    Source ↗
    A 80.8%B 76.7%
    Winner: Claude Opus 4.6Δ 4.1
    SWE-bench Verified: Claude Opus 4.6 scored 80.8%; Grok 4.20 scored 76.7%. Claude Opus 4.6 wins this benchmark.

  3. MMMU-Pro

    Multimodal
    Source ↗
    A 77.3%B 75.2%
    Winner: Claude Opus 4.6Δ 2.1
    MMMU-Pro: Claude Opus 4.6 scored 77.3%; Grok 4.20 scored 75.2%. Claude Opus 4.6 wins this benchmark.

  4. SWE-bench Pro

    Coding
    Source ↗
    A 53.4%B 51.8%
    Winner: Claude Opus 4.6Δ 1.6
    SWE-bench Pro: Claude Opus 4.6 scored 53.4%; Grok 4.20 scored 51.8%. Claude Opus 4.6 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricClaude Opus 4.6Grok 4.20Comparison
Input / output priceUSD per 1M tokensClaude Opus 4.6$5 input / $25 outputGrok 4.20$2 input / $6 outputGrok 4.20 has the lower combined listed price.
Generation speedtokens per secondClaude Opus 4.640 tok/sGrok 4.20233 tok/sGrok 4.20 has the higher measured throughput.
First-answer latencyseconds to first tokenClaude Opus 4.61.78 sGrok 4.2010.33 sClaude Opus 4.6 reaches the first token sooner.
Context windowmaximum listed tokensClaude Opus 4.61MGrok 4.202MGrok 4.20 lists the larger context window.

Benchmark Deep Dive

AgenticClaude Opus 4.6 wins
BenchmarkClaude Opus 4.6Grok 4.20Result
Terminal-Bench 2.0Source 65.4%47.1%Claude Opus 4.6 leads
BrowseCompSource 83.7%—Not comparable
OSWorld-VerifiedSource 72.7%—Not comparable
τ²-bench resultsSource 84.8%—Not comparable
Claw-EvalSource 70.4%—Not comparable
DeepSearchQASource 73.7%62.8%Claude Opus 4.6 leads
CyberGymSource 66.6%—Not comparable
Gert LabsSource 61.85%38.36%Claude Opus 4.6 leads
ResearchClawBenchSource 19.9%—Not comparable
JobBenchSource 36.7%—Not comparable
CodingClaude Opus 4.6 wins
BenchmarkClaude Opus 4.6Grok 4.20Result
SWE-bench VerifiedSource 80.8%76.7%Claude Opus 4.6 leads
SWE-bench Verified*Source 75.6%—Not comparable
LiveCodeBench ProSource 70.7%74.2%Grok 4.20 leads
SWE-bench ProSource 53.4%51.8%Claude Opus 4.6 leads
SWE-RebenchSource 65.3%—Not comparable
React Native EvalsSource 84.1%—Not comparable
Vibe Code BenchSource 57.57%4.06%Claude Opus 4.6 leads
AA-SciCodeSource 45.7%—Not comparable
FrontierCode 1.1 MainSource 26.9%—Not comparable
Reasoning
BenchmarkClaude Opus 4.6Grok 4.20Result
AA-LCRSource 58.3%—Not comparable
CritPtSource 2.8%—Not comparable
ARC-AGI-2Source —53.3%Not comparable
Knowledge
BenchmarkClaude Opus 4.6Grok 4.20Result
GPQASource 91.3%—Not comparable
GPQA-DSource 89.2%88.5%Claude Opus 4.6 leads
SuperGPQASource 95%—Not comparable
MMLU-ProSource 82%—Not comparable
MMLU-Pro (Arcee)Source 89.1%—Not comparable
HLESource 53%—Not comparable
HLE w/o toolsSource 40%31.6%Claude Opus 4.6 leads
HealthBench HardSource 14.8%20.3%Grok 4.20 leads
MedXpertQA (Text)Source 52.1%50.2%Claude Opus 4.6 leads
Artificial Analysis Intelligence IndexSource 37.8%—Not comparable
AA-GPQA DiamondSource 84.0%—Not comparable
AA-HLESource 18.6%—Not comparable
AA-Omniscience IndexSource 3.5%—Not comparable
AA-Omniscience AccuracySource 45.2%—Not comparable
AA-Omniscience Hallucination RateSource 76.0%—Not comparable
Math
BenchmarkClaude Opus 4.6Grok 4.20Result
AIME25 (Arcee)Source 99.8%—Not comparable
FrontierMath v2 (Tiers 1-3)Source 40.700%—Not comparable
FrontierMath v2 (Tier 4)Source 22.900%—Not comparable
MultimodalClaude Opus 4.6 wins
BenchmarkClaude Opus 4.6Grok 4.20Result
MMMU-ProSource 77.3%75.2%Claude Opus 4.6 leads
ERQASource 51.6%54.1%Grok 4.20 leads
ScreenSpot ProSource 83.1%—Not comparable
MedXpertQA (MM)Source 64.8%65.8%Grok 4.20 leads
AA-MMMU-ProSource 72.5%—Not comparable
Design Arena WebsiteSource 13251257Claude Opus 4.6 leads
CharXivSource —60.9%Not comparable
SimpleVQASource —57.4%Not comparable
Inst. Following
BenchmarkClaude Opus 4.6Grok 4.20Result
AA-IFBenchSource 44.6%—Not comparable
Frequently Asked Questions (4)

Which is better, Claude Opus 4.6 or Grok 4.20?

Claude Opus 4.6 is ahead on BenchLM's BenchAlign leaderboard, 68.59 to 54.68. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 65.4% and 47.1%.

Which is better for coding, Claude Opus 4.6 or Grok 4.20?

Claude Opus 4.6 has the edge for coding in this comparison, averaging 68.1 versus 67.1. Inside this category, Vibe Code Bench is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Claude Opus 4.6 or Grok 4.20?

Claude Opus 4.6 has the edge for agentic tasks in this comparison, averaging 73 versus 47.1. Inside this category, Gert Labs is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Claude Opus 4.6 or Grok 4.20?

Claude Opus 4.6 has the edge for multimodal and grounded tasks in this comparison, averaging 77.3 versus 70.1. Inside this category, Design Arena Website is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Anthropic Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 23, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.