Claude Sonnet 4.5 vs Grok 4
Wins 2 of 8 areas
Long documents · Following instructions
Wins 5 of 8 areas
Coding · Agents · Reasoning · Facts · Math
Grok 4 is the stronger all-rounder.Claude Sonnet 4.5 is better at long documents.
Scores updated · 34 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.
- ReasoningHard problems that need careful thinking03Grok 43 of 4 tests · 1 tie
- FactsGetting facts right instead of making them up12Grok 42 of 3 tests
- CodingWriting and fixing software01Grok 41 of 2 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own01Grok 41 of 1 test
- MathCompetition and research-level math01Grok 41 of 1 test
- Long documentsFinding answers in very long texts10Claude Sonnet 4.51 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Sonnet 4.51 of 1 test
- Images and chartsUnderstanding pictures, charts and video00Even0 each · 1 tie
Agents, images and charts, math, 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.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Grok 4 pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+17.2points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+8.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+7.6points ahead
Where Claude Sonnet 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+15.4points ahead
- Reasons across sets of long documentsAA-LCR+4.3points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+3.6points ahead
Every test, side by side
All 34 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningGrok 4
- Humanity's Last ExamGrok 4 by 8.917.826.7+8.9
- GPQA DiamondGrok 4 by 4.383.487.7+4.3
- ARC-AGI-2Grok 4 by 2.313.615.9+2.3
- CritPttie1.12tie
FactsGrok 4
- SimpleQA VerifiedGrok 4 by 17.230.747.9+17.2
- AA-Omniscience · Non-hallucinationClaude Sonnet 4.5 by 15.450.935.5+15.4
- AA-Omniscience · AccuracyGrok 4 by 7.632.940.5+7.6
Following instructionsClaude Sonnet 4.5
- IFBenchClaude Sonnet 4.5 by 3.657.353.7+3.6
Other results20 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.
- HarmBenchClaude Sonnet 4.5 by 52.39239.7+52.3
- AIR-Bench 2024Claude Sonnet 4.5 by 45.489.844.4+45.4
- AA Agentic IndexGrok 4 by 2417.541.5+24
- HMMT 2025Grok 4 by 15.474.690+15.4
- HMMT25 no toolsGrok 4 by 15.474.690+15.4
- Artificial Analysis Coding IndexClaude Sonnet 4.5 by 11.652.140.5+11.6
- LiveCodeBenchGrok 4 by 87179+8
- simple_safety_testsClaude Sonnet 4.5 by 7.810092.2+7.8
- HMMT Nov. 2025Grok 4 by 6.681.788.3+6.6
- matharena_visual_math_overallClaude Sonnet 4.5 by 5.875.870+5.8
- bbqClaude Sonnet 4.5 by 5.298.993.7+5.2
- AIME 2025Grok 4 by 4.78791.7+4.7
- AIME25 no toolsGrok 4 by 3.78891.7+3.7
- τ²-Bench Telecom (AA run)Claude Sonnet 4.5 by 3.278.174.9+3.2
- ARC-AGI-1Grok 4 by 363.766.7+3
- anthropic_red_teamClaude Sonnet 4.5 by 2.598.295.7+2.5
- AA-OmniscienceGrok 4 by 2.2-0.12.1+2.2
- frontiermath_tier_4_v1Claude Sonnet 4.5 by 2.14.22.1+2.1
- AA IntelligenceGrok 4 by 1.820.722.5+1.8
- XSTesttie96.496.6tie
Questions people ask
Which is better, Claude Sonnet 4.5 or Grok 4?
Grok 4 wins five of the eight areas we test: coding, agents, reasoning, facts and math. Claude Sonnet 4.5 wins long documents and following instructions. They are level on images and charts.
Which is better for coding?
Grok 4. It wins 1 of the 2 coding tests both models report; Claude Sonnet 4.5 wins none, and 1 is a tie.
Which is cheaper?
Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens; Grok 4 costs $3.00 and $15.00. For a million tokens read plus a million written, they cost about the same.
How do you compare the two?
We use the 34 benchmark tests both models have published scores on. The verdict counts the 14 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.