Anthropic: Claude Sonnet 4.6

Benchmarks gegenereerd uit Aibenchy-testsuites op : 19 feb 2026

Categorie-uitsplitsing

Categorie	Volledig geslaagde tests	Score	Consistentie	Slaagpercentage per poging	Instabiele tests	Redeneerscore	Kosten
Anti-AI Tricks	0/2	1.00	10.00	0.0%	0	-	$0.01092
Data parsing and extraction	2/2	10.00	10.00	100.0%	0	-	$0.02854
Domain specific	2/3	7.00	10.00	66.7%	0	-	$0.00309
Instructions following	1/2	5.50	10.00	50.0%	0	-	$0.00342
Puzzle Solving	1/3	5.00	7.68	44.4%	1	-	$0.01092

Vergelijk Anthropic: Claude Sonnet 4.6 met...

#12 · OpenAI

Redenering (medium)

Score: 5.75

Consistentie: 7.19

Slaagpercentage per poging: 63.9%

Instabiele tests: 4

Kosten per resultaat: 0.0951

Correcte tests: 6/12

Totale kosten: $0.00571

#14 · Qwen

Zonder redenering

Score: 5.67

Consistentie: 9.99

Slaagpercentage per poging: 50.0%

Instabiele tests: 0

Kosten per resultaat: 0.0997

Correcte tests: 6/12

Totale kosten: $0.00599

#11 · OpenAI

Redenering (medium)

Score: 5.92

Consistentie: 6.03

Slaagpercentage per poging: 72.2%

Instabiele tests: 6

Kosten per resultaat: 0.4675

Correcte tests: 6/12

Totale kosten: $0.02806

Vergelijk Anthropic: Claude Sonnet 4.6 met...