Claude Sonnet 5.5 vs GPT-6.1 Sol
Claude Sonnet 5.5 is ahead on some work, and they tie on the rest. GPT-6.1 Sol is the cheaper one, by 2.6×.
GrindBench-PubSec
BClaude Sonnet 5.5Anthropic70% 64–75%3.5¢ a task · 300 tasksCGPT-6.1 SolOpenAI58% 52–63%1.3¢ a task · 300 tasks
Claude Sonnet 5.5 is ahead here by 12 points, more than chance explains.
| Kind of work | Claude Sonnet 5.5 | GPT-6.1 Sol |
|---|---|---|
| ATIP redaction | 25% | 0% |
| Procurement evaluation | 90% | 80% |
| Performance analysis | 72% | 60% |
| Bilingual equivalence | 85% | 85% |
GrindBench-Banking
AClaude Sonnet 5.5Anthropic80% 70–90%$0.14 a task · 60 tasksBGPT-6.1 SolOpenAI63% 50–75%5.4¢ a task · 60 tasks
Too close to call here: their ranges overlap, so on another set of tasks the order could flip.
| Kind of work | Claude Sonnet 5.5 | GPT-6.1 Sol |
|---|---|---|
| Lending & mortgages | 80% | 60% |
| AML & KYC | 80% | 30% |
| Deposits & payments | 60% | 60% |
| Complaints & conduct | 90% | 100% |
| Credit risk | 80% | 60% |
| Capital & liquidity | 90% | 70% |
Each kind of work here rests on 10 tasks per model, so a single row can easily be 20 points off. A lead of 10 points or more is in bold; read it as a hint, not a result.