GPT-6 Sol versus Claude Opus 5.5 on developer-test results and cost
The New Stack reports that Claude Opus 5.5 was perfect on all 15 runs in its developer tests, while GPT-6 Sol was perfect on 12 at about a sixth of the cost.
TLDR
The New Stack says it ran both models through three hard developer tests five times each. It reports that Claude Opus 5.5 was perfect on all 15 runs and GPT-6 Sol on 12, with Sol costing about a sixth as much.
Combined views
474
1 Source, first seen 4h ago
1 likes