Claude Sonnet 4 vs Grok 4 Fast
Wins 2 of 7 areas
Facts · Following instructions
Wins 3 of 7 areas
Agents · Reasoning · Long documents
Grok 4 Fast wins more areas, narrowly.Claude Sonnet 4 is better at facts.
Scores updated · 28 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. 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 4 Fast3 of 4 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own02Grok 4 Fast2 of 2 tests
- Long documentsFinding answers in very long texts01Grok 4 Fast1 of 1 test
- FactsGetting facts right instead of making them up20Claude Sonnet 42 of 3 tests · 1 tie
- Following instructionsDoing exactly what it is asked10Claude Sonnet 41 of 1 test
- CodingWriting and fixing software11Even1 each
- Images and chartsUnderstanding pictures, charts and video00Even0 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 Fast costs 96% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Grok 4 Fast pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+32.7points ahead
- Real work tasks from 44 professionsGDPVal+16points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+8.4points ahead
Where Claude Sonnet 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+39.2points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+12.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+4.2points ahead
Every test, side by side
All 28 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- Terminal-Bench HardClaude Sonnet 4 by 12.231.118.9+12.2
- SciCodeGrok 4 Fast by 4.24044.2+4.2
AgentsGrok 4 Fast
- BrowseCompGrok 4 Fast by 32.712.244.9+32.7
- GDPValGrok 4 Fast by 169.725.7+16
ReasoningGrok 4 Fast
- Humanity's Last ExamGrok 4 Fast by 8.410.719.1+8.4
- GPQA DiamondGrok 4 Fast by 777.784.7+7
- CritPtGrok 4 Fast by 2.60.32.9+2.6
- ARC-AGI-2tie5.95.3tie
FactsClaude Sonnet 4
- AA-Omniscience · Non-hallucinationClaude Sonnet 4 by 39.270.931.7+39.2
- Vectara HHEM hallucination ratelower is betterClaude Sonnet 4 by 9.910.320.2+9.9
- AA-Omniscience · Accuracytie22.722.8tie
Following instructionsClaude Sonnet 4
- IFBenchClaude Sonnet 4 by 4.254.750.5+4.2
Other results14 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.
- SimpleQAGrok 4 Fast by 79.115.995+79.1
- HMMT 2025Grok 4 Fast by 77.415.993.3+77.4
- AA-OmniscienceClaude Sonnet 4 by 30.10.2-29.9+30.1
- AA Agentic IndexGrok 4 Fast by 21.917.639.5+21.9
- AIME 2025Grok 4 Fast by 21.570.592+21.5
- BrowseComp-ZHGrok 4 Fast by 15.929.145+15.9
- LiveCodeBenchGrok 4 Fast by 11.568.580+11.5
- Artificial Analysis Coding IndexClaude Sonnet 4 by 10.237.627.4+10.2
- vectara_factual_consistencyClaude Sonnet 4 by 9.989.779.8+9.9
- ARC-AGI-1Grok 4 Fast by 8.54048.5+8.5
- vectara_avg_summary_lengthGrok 4 Fast by 28.1 rating points145.8173.9+28.1 rating
- τ²-Bench Telecom (AA run)Grok 4 Fast by 1.264.665.8+1.2
- AA IntelligenceClaude Sonnet 4 by 118.917.9+1
- vectara_answer_ratetie98.699.5tie
Questions people ask
Which is better, Claude Sonnet 4 or Grok 4 Fast?
Grok 4 Fast wins three of the seven areas where both have results: agents, reasoning and long documents. Claude Sonnet 4 wins facts and following instructions. They are level on coding and images and charts.
Which is better for coding?
Neither. They win 1 coding test each of the 2 both models report.
Which is cheaper?
Claude Sonnet 4 costs $3.00 per million input tokens and $15.00 per million output tokens; Grok 4 Fast costs $0.20 and $0.50. That makes Grok 4 Fast about 96% cheaper for the same work.
How do you compare the two?
We use the 28 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 14 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.