Do more expensive AI coding models write fewer bugs?

As of Nov 9, 2026, the most expensive models, Claude Fable 5.1 and GPT-6 Astra ($50 per 1M output tokens), had relative bug rates of 0.88 and 0.95. The lowest rate belonged to GPT-6 Luna (0.31, $0.50 per 1M output tokens).

Updated Nov 9, 2026

AI coding model bug rate vs price

As of Nov 9, 2026, GPT-6 Luna had the lowest relative bug rate (0.31, $0.50 per 1M output tokens), and was also the cheapest model with a below-average bug rate.

Lower is better

Rates and list prices

USD per 1M tokens

Ranked models by relative bug rate with input and output list prices per 1M tokens.
Model	Rate	Input	Output
GPT-6 Luna	0.31	$0.10	$0.50
GPT-6.1 Sol	0.74	$2	$10
Claude Fable 5.1	0.88	$10	$50
GPT-6 Astra	0.95	$10	$50
Claude Sonnet 5.5	1.04	$2	$10
Claude Opus 5.5	1.11	$4	$20
Grok 4.7	1.61	$2	$6

Questions and answers

Do more expensive AI coding models write fewer bugs?: As of Nov 9, 2026, the most expensive models, Claude Fable 5.1 and GPT-6 Astra ($50 per 1M output tokens), had relative bug rates of 0.88 and 0.95. The lowest rate belonged to GPT-6 Luna (0.31, $0.50 per 1M output tokens).