Claude Haiku 4.5 vs Claude Sonnet 4.6
Wins 0 of 7 areas
—
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Images and charts · Long documents · Following instructions
Claude Sonnet 4.6 is the stronger all-rounder.Claude Haiku 4.5 is cheaper.
Scores updated · 36 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.
- CodingWriting and fixing software06Claude Sonnet 4.66 of 6 tests
- AgentsCarrying out multi-step tasks on its own05Claude Sonnet 4.65 of 5 tests
- ReasoningHard problems that need careful thinking04Claude Sonnet 4.64 of 4 tests
- FactsGetting facts right instead of making them up12Claude Sonnet 4.62 of 4 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video01Claude Sonnet 4.61 of 1 test
- Long documentsFinding answers in very long texts01Claude Sonnet 4.61 of 1 test
- Following instructionsDoing exactly what it is asked01Claude Sonnet 4.61 of 1 test
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.
Claude Haiku 4.5 costs 67% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Claude Sonnet 4.6 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+54.3points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+27points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+25.7points ahead
Where Claude Haiku 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+21.1points ahead
Every test, side by side
All 36 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 2744.271.2+27
- Terminal-Bench HardClaude Sonnet 4.6 by 25.727.353+25.7
- LMArena · WebDevClaude Sonnet 4.6 by 192 rating points13291521+192 rating
- SWE-bench ProClaude Sonnet 4.6 by 18.639.558.1+18.6
- SciCodeClaude Sonnet 4.6 by 7.942.250.1+7.9
- SWE-bench VerifiedClaude Sonnet 4.6 by 6.373.379.6+6.3
AgentsClaude Sonnet 4.6
- τ-Bench V3 · BankingClaude Sonnet 4.6 by 25.19.334.4+25.1
- GDPValClaude Sonnet 4.6 by 24.911.836.7+24.9
- OSWorld-VerifiedClaude Sonnet 4.6 by 21.850.772.5+21.8
- AA IT-Bench SREClaude Sonnet 4.6 by 12.527.339.8+12.5
- Terminal-Bench 4.0Claude Sonnet 4.6 by 303+3
ReasoningClaude Sonnet 4.6
- ARC-AGI-2Claude Sonnet 4.6 by 54.3458.3+54.3
- Humanity's Last ExamClaude Sonnet 4.6 by 23.210.433.6+23.2
- GPQA DiamondClaude Sonnet 4.6 by 20.367.287.5+20.3
- CritPtClaude Sonnet 4.6 by 3.103.1+3.1
FactsClaude Sonnet 4.6
- AA-Omniscience · AccuracyClaude Sonnet 4.6 by 22.91840.9+22.9
- AA-Omniscience · Non-hallucinationClaude Haiku 4.5 by 21.172.751.6+21.1
- SimpleQA VerifiedClaude Sonnet 4.6 by 19.613.232.8+19.6
- Vectara HHEM hallucination ratelower is bettertie9.810.6tie
Images and chartsClaude Sonnet 4.6
- MMMU-ProClaude Sonnet 4.6 by 14.758.673.3+14.7
Following instructionsClaude Sonnet 4.6
- IFBenchClaude Sonnet 4.6 by 2.354.356.6+2.3
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.
- ARC-AGI-1Claude Sonnet 4.6 by 38.347.786+38.3
- AA Agentic IndexClaude Sonnet 4.6 by 22.810.333.1+22.8
- τ²-Bench Telecom (AA run)Claude Sonnet 4.6 by 2154.775.7+21
- Finance Agent v2Claude Sonnet 4.6 by 203151+20
- Artificial Analysis Coding IndexClaude Sonnet 4.6 by 19.143.963+19.1
- AA-OmniscienceClaude Sonnet 4.6 by 16.6-4.412.2+16.6
- AA IntelligenceClaude Sonnet 4.6 by 13.216.930.1+13.2
- τ²-Bench (Retail)Claude Sonnet 4.6 by 8.583.291.7+8.5
- MMMLUClaude Sonnet 4.6 by 6.38389.3+6.3
- frontiermath_tier_4_v1Claude Sonnet 4.6 by 6.22.18.3+6.2
- vectara_factual_consistencytie90.289.4tie
- FLTEvaltie2323.7tie
- vectara_answer_ratetie99.599.9tie
- vectara_avg_summary_lengthtie115.1114.7tie
Questions people ask
Which is better, Claude Haiku 4.5 or Claude Sonnet 4.6?
Claude Sonnet 4.6 wins all seven areas where both have results: coding, agents, reasoning, facts, images and charts, long documents and following instructions. Claude Haiku 4.5 wins none, but costs 67% less.
Which is better for coding?
Claude Sonnet 4.6. It wins 6 of the 6 coding tests both models report; Claude Haiku 4.5 wins none.
Which is cheaper?
Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens; Claude Sonnet 4.6 costs $3.00 and $15.00. That makes Claude Haiku 4.5 about 67% cheaper for the same work.
How do you compare the two?
We use the 36 benchmark tests both models have published scores on. The verdict counts the 22 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.