GPT-5.6 Sol vs Claude Fable 5: 59 AI Index, $1.04/task

OpenAI's GPT-5.6 Sol scored 59 points on the Artificial Analysis Intelligence Index, placing it just one point behind Claude Fable 5. At $1.04 per task, Sol costs a third of what Anthropic's top model charges, and in agentic coding Sol beats every competitor.

How did GPT-5.6 Sol perform on benchmarks?

GPT-5.6 Sol scored 59 points on the Artificial Analysis Intelligence Index, one point behind Claude Fable 5, which places Sol nearly level with Fable 5 on aggregated benchmarks. That single-point gap frames Sol as almost parity on the index while still trailing Claude Fable 5 by a narrow margin.

The headline figure is an aggregated measure; the source cites the Artificial Analysis Intelligence Index as the benchmark used and gives Sol's result as 59 points and the gap to Claude Fable 5 as one point. The coverage frames that aggregated score as the central comparison between OpenAI's Sol and Claude Fable 5.

How does Sol compare on price and agentic coding?

Sol costs $1.04 per task, which the source says is a third of what Anthropic's top model charges, and Sol "beats every competitor" in agentic coding. The two facts together present Sol as both a price and performance pressure point: lower per-task cost and a leading position in agentic coding workloads.

The underlying reporting links the $1.04 per task price directly to GPT-5.6 Sol and states the ratio to Anthropic's top model rather than a second exact price. On capability, the source explicitly states Sol beats every competitor in agentic coding, positioning that capability as a notable strength alongside the cost advantage.

Why it matters

A model that scores 59 on an aggregated index while costing a third of a rival's top offering reshapes competitive incentives: buyers weighing benchmark parity and price now see Sol as an option that narrows the performance gap while undercutting cost. The source also notes that Sol's agentic coding lead increases pricing pressure on Anthropic.

Put simply, the combination of near-parity on an aggregated index and a substantially lower stated per-task price forces competitors to answer on both fronts: close the narrow benchmark gap or match the lower cost. The source highlights pricing pressure on Anthropic as the immediate industry implication.

What to watch

Watch whether Anthropic responds on price, capability, or both: the coverage cites Sol's $1.04 per task and its agentic coding advantage as the two concrete levers applying pressure. A follow-up signal would be any change in Anthropic's pricing or public benchmark results that closes the one-point gap cited here.

If Anthropic lowers price points for its top model or publishes new benchmark numbers that eliminate the one-point difference on the Artificial Analysis Intelligence Index, that would confirm a direct competitive reaction to Sol's combined price and capability positioning.