今天新增gpt-6IMAGE
DeepFinance 基准：GPT-6 Astra 最高思考档 57.6% 登顶
团队发布 DeepFinance 长周期金融任务基准，发现更多思考确实有效；GPT-6 Astra 在最高思考档以 57.6% 领先 Opus 5.5 的 54.8%。
rahul
@0interestrates3 小时前收录原推发布于 2026.11.08
rahul @0interestrates原推发布于 2026.11.08
after releasing DeepFinance bench, we wanted to see whether more thinking helps on ultra long-horizon finance tasks or if it just added “token-noise”. turns out, it helps. GPT-6 Astra on max thinking is the new leader on DeepFinance at 57.6%, exceeding Claude Opus 5.5 (54.8%) andClaude Sonnet 5.5 (54.6 %)
这条是引用着另一条写的：x.com/0interestrates/status/2118473925902635507
x.com/0interestrates/status/2119243014011019462采集于 2026.11.08 10:09