Claude Sonnet 4.6 vs Grok 4.3
Wins 4 of 8 areas
Coding · Agents · Math · Long documents
Wins 2 of 8 areas
Reasoning · Following instructions
Claude Sonnet 4.6 wins more areas, narrowly.Grok 4.3 is cheaper and better at reasoning.
Scores updated · 31 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.
- CodingWriting and fixing software60Claude Sonnet 4.66 of 6 tests
- AgentsCarrying out multi-step tasks on its own50Claude Sonnet 4.65 of 5 tests
- MathCompetition and research-level math10Claude Sonnet 4.61 of 1 test
- Long documentsFinding answers in very long texts10Claude Sonnet 4.61 of 1 test
- ReasoningHard problems that need careful thinking13Grok 4.33 of 4 tests
- Following instructionsDoing exactly what it is asked01Grok 4.31 of 2 tests · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
- Images and chartsUnderstanding pictures, charts and video11Even1 each
Math and long documents 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 4.3 costs 79% 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 Sonnet 4.6 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+31.5points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+24.1points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+22points ahead
Where Grok 4.3 pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+24.7points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+22.6points ahead
- Research-level physics problemsCritPt+4.9points ahead
Every test, side by side
All 31 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4.6
- Terminal-Bench 2.1Claude Sonnet 4.6 by 31.571.239.7+31.5
- LiveBench · Agentic CodingClaude Sonnet 4.6 by 24.142.618.5+24.1
- LMArena · WebDevClaude Sonnet 4.6 by 165 rating points15221357+165 rating
- Terminal-Bench HardClaude Sonnet 4.6 by 15.15337.9+15.1
- LiveBench · CodingClaude Sonnet 4.6 by 9.479.369.9+9.4
- SciCodeClaude Sonnet 4.6 by 1.850.148.3+1.8
AgentsClaude Sonnet 4.6
- τ-Bench V3 · BankingClaude Sonnet 4.6 by 2234.412.4+22
- GDPValClaude Sonnet 4.6 by 14.736.722+14.7
- AA ApexAgentsClaude Sonnet 4.6 by 112817+11
- AA IT-Bench SREClaude Sonnet 4.6 by 7.139.832.7+7.1
- Terminal-Bench 4.0Claude Sonnet 4.6 by 330+3
ReasoningGrok 4.3
- LiveBench · ReasoningClaude Sonnet 4.6 by 1484.870.8+14
- CritPtGrok 4.3 by 4.93.18+4.9
- Humanity's Last ExamGrok 4.3 by 3.633.637.2+3.6
- GPQA DiamondGrok 4.3 by 2.687.590.1+2.6
FactsEven
- AA-Omniscience · Non-hallucinationGrok 4.3 by 22.651.674.2+22.6
- AA-Omniscience · AccuracyClaude Sonnet 4.6 by 6.140.934.8+6.1
Images and chartsEven
- LMArena · VisionClaude Sonnet 4.6 by 54 rating points12831229+54 rating
- MMMU-ProGrok 4.3 by 4.873.378.1+4.8
Following instructionsGrok 4.3
- IFBenchGrok 4.3 by 24.756.681.3+24.7
- LiveBench · Instruction Followingtie63.262.8tie
Other results8 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.
- livebench_data_analysisClaude Sonnet 4.6 by 22.27855.8+22.2
- τ²-Bench Telecom (AA run)Grok 4.3 by 2275.797.7+22
- Artificial Analysis Coding IndexClaude Sonnet 4.6 by 20.86342.2+20.8
- AA Agentic IndexClaude Sonnet 4.6 by 15.933.117.2+15.9
- Finance Agent v2Claude Sonnet 4.6 by 13.35137.7+13.3
- AA-OmniscienceGrok 4.3 by 5.812.218+5.8
- AA IntelligenceClaude Sonnet 4.6 by 5.230.124.9+5.2
- livebench_languageClaude Sonnet 4.6 by 2.576.173.6+2.5
Questions people ask
Which is better, Claude Sonnet 4.6 or Grok 4.3?
Claude Sonnet 4.6 wins four of the eight areas we test: coding, agents, math and long documents. Grok 4.3 wins reasoning and following instructions, and costs 79% less. They are level on facts and images and charts.
Which is better for coding?
Claude Sonnet 4.6. It wins 6 of the 6 coding tests both models report; Grok 4.3 wins none.
Which is cheaper?
Claude Sonnet 4.6 costs $3.00 per million input tokens and $15.00 per million output tokens; Grok 4.3 costs $1.25 and $2.50. That makes Grok 4.3 about 79% cheaper for the same work.
How do you compare the two?
We use the 31 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 8 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.