Claude Opus 4.8 vs Grok 4.6
Wins 3 of 8 areas
Facts · Images and charts · Math
Wins 3 of 8 areas
Coding · Agents · Long documents
The two are evenly matched.Claude Opus 4.8 is better at math and facts; Grok 4.6 at agents and coding.
Scores updated · 35 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- MathCompetition and research-level math30Claude Opus 4.83 of 3 tests
- FactsGetting facts right instead of making them up21Claude Opus 4.82 of 3 tests
- Images and chartsUnderstanding pictures, charts and video10Claude Opus 4.81 of 1 test
- AgentsCarrying out multi-step tasks on its own13Grok 4.63 of 4 tests
- CodingWriting and fixing software23Grok 4.63 of 5 tests
- Long documentsFinding answers in very long texts01Grok 4.61 of 1 test
- ReasoningHard problems that need careful thinking33Even3 each · 1 tie
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Images and charts, long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
Grok 4.6 costs 73% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Claude Opus 4.8 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+24.4points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+14points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+5.8points ahead
Where Grok 4.6 pulls ahead
- Long professional tasks in banking, consulting and lawAA ApexAgents+18.1points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+15.3points ahead
- Common-sense trick questionsSimpleBench+11.1points ahead
Every test, side by side
All 35 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGrok 4.6
- LiveBench · Agentic CodingGrok 4.6 by 6.550.557+6.5
- LMArena · WebDevGrok 4.6 by 61 rating points15561617+61 rating
- LiveBench · CodingClaude Opus 4.8 by 581.876.8+5
- Terminal-Bench 2.1Grok 4.6 by 3.484.688+3.4
- SciCodeClaude Opus 4.8 by 1.454.453+1.4
AgentsGrok 4.6
- AA ApexAgentsGrok 4.6 by 18.139.457.5+18.1
- GDPValGrok 4.6 by 9.547.857.3+9.5
- τ-Bench V3 · BankingGrok 4.6 by 9.134.243.3+9.1
- Terminal-Bench 4.0Claude Opus 4.8 by 4.521.717.2+4.5
ReasoningEven
- SimpleBenchGrok 4.6 by 11.164.875.9+11.1
- ARC-AGI-2Claude Opus 4.8 by 572.167.1+5
- Humanity's Last ExamClaude Opus 4.8 by 4.648.744.1+4.6
- GPQA DiamondGrok 4.6 by 1.59293.5+1.5
- LiveBench · ReasoningGrok 4.6 by 1.389.290.5+1.3
- CritPtClaude Opus 4.8 by 1.220.919.7+1.2
- ARC-AGI-3tie1.52.1tie
FactsClaude Opus 4.8
- AA-Omniscience · Non-hallucinationGrok 4.6 by 15.360.776+15.3
- AA-Omniscience · AccuracyClaude Opus 4.8 by 5.848.843+5.8
- SimpleQA VerifiedClaude Opus 4.8 by 4.15348.9+4.1
Images and chartsClaude Opus 4.8
- LMArena · VisionClaude Opus 4.8 by 30 rating points12951265+30 rating
MathClaude Opus 4.8
- FrontierMath Tier 4Claude Opus 4.8 by 24.456.131.7+24.4
- FrontierMath Tiers 1-3 (v2)Claude Opus 4.8 by 148066+14
- LiveBench · MathematicsClaude Opus 4.8 by 1.794.392.6+1.7
Following instructionsEven
- LiveBench · Instruction Followingtie7271.9tie
Other results10 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- AA Agentic IndexGrok 4.6 by 10.142.652.7+10.1
- livebench_data_analysisGrok 4.6 by 7.96673.9+7.9
- DeepSWE 1.1Grok 4.6 by 6.95965.9+6.9
- ARC-AGI-1Claude Opus 4.8 by 5.592.587+5.5
- Terminal-bench 3.0Grok 4.6 by 4.921.126+4.9
- livebench_languageGrok 4.6 by 479.783.7+4
- Terminal-Bench 4.0Claude Opus 4.8 by 3.323.620.3+3.3
- AA IntelligenceGrok 4.6 by 2.441.844.2+2.4
- Artificial Analysis Coding IndexGrok 4.6 by 1.674.375.9+1.6
- AA-Omnisciencetie28.829.3tie
Questions people ask
Which is better, Claude Opus 4.8 or Grok 4.6?
Claude Opus 4.8 and Grok 4.6 each win three of the eight areas we test. Claude Opus 4.8 wins facts, images and charts and math; Grok 4.6 wins coding, agents and long documents. Grok 4.6 costs 73% less. They are level on reasoning and following instructions.
Which is better for coding?
Grok 4.6. It wins 3 of the 5 coding tests both models report; Claude Opus 4.8 wins 2.
Which is cheaper?
Claude Opus 4.8 costs $5.00 per million input tokens and $25.00 per million output tokens; Grok 4.6 costs $2.00 and $6.00. That makes Grok 4.6 about 73% cheaper for the same work.
How do you compare the two?
We use the 35 benchmark tests both models have published scores on. The verdict counts the 25 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.