以下部分引用 AI 总结，现阶段 AI 仍然有幻觉。请以内容的原文为准。
数据新鲜度群聊 07-14 ✓卖方 07-14 ✓Wrap 断档15天 ✕News 07-15 ✓X 断档12天 ✕页面生成 07-15
GPT 5.5 Pro
12 atoms · 跨 11 天 · 首见 2026-04-25 · 最近 2026-06-27
三色: 🟦 fact 6 · 🟥 take 6 · stance ▲2/▼1/◆3
来源: X 12
时态: fresh:10 · aging:2
标签: 好观点:5 · 好数字:4 · 好信源:2 · 好问题:1 · 好思考:1
叙事 (narrative) (6)
- 2026-06-14 · 技能文件编写能力评估
query gpt 5. 5 pro, sanity check, and ask it to rewrite or clarify or suggest vague new strategies. The results are not very good
tags: 好观点 · → daily
value: direction=差
- 2026-06-11 · 模型质量极高
pro is an insanely good model that barely makes mistakes
tags: 好观点 · → daily
- 2026-05-24 · 能力水平
GPT 5.5 Pro 是高于 xhigh 的级别，能构建其他模型声称困难或不可能的内容
tags: 好观点 · → daily
value: direction=高于 xhigh
- 2026-05-23 · 是否为独立模型
GPT 5.5 Pro 是 5.5 xhigh 集成或测试时计算扩展变体，而非独立模型
tags: 好观点 · → daily
value: direction=否定
- 2026-05-11
[老化中] · 被描述为 close to Stockfish level 7
GPT 5.5 pro is close to Stockfish level 7
tags: 好观点 · → daily
- 2026-04-30
[老化中] · 推理模式选择（正常/扩展）
Do you use normal or extended thinking for GPT 5.5 Pro
tags: 好问题 · → daily
事实 (fact) (6)
- 2026-06-27 · 用于解决9个挑战性开放问题的模型
We design a simple pipeline (using GPT 5.5 Pro and Claude Opus 4.8) that resolves 9 challenging open problems, including open problems from prominent theoretical computer science venues—4 from COLT open problem list and 1 from FOCS —as well as 4 problems from the commutative algebra.
tags: 好信源·好数字 · → daily
- 2026-05-05 · can generate similar paper
I can just ask GPT 5.5 Pro or DeepResearch to generate this
tags: 好信源 · → daily
- 2026-06-09 · price per M input tokens
AND it's about the same price as GPT 5.5 ($10/M input, $45/M output) vs Fable 5 ($10/M input, $50/M output) and 6x cheaper than GPT 5.5 Pro.
tags: 好数字 · → daily
value: qty=6x more expensive than GPT 5.5
- 2026-05-21 · 未受益于physics-intern harness
GPT 5.5 Pro actually didn't benefit from the physics-intern harness
tags: 好思考 · → daily
value: date=2026-05-21
- 2026-04-25 · Mensa Norway score
GPT 5.5 Pro vision is actually the first model to score 145, on the Mensa Norway test
tags: 好数字 · → daily
value: qty=145
- 2026-04-25 · Mensa Norway average score
It ended up averaging 144 so just below GPT 5.5 Pro
tags: 好数字 · → daily
value: qty=144
时间轴 (近 20)
- 2026-06-27 ·
fact · 用于解决9个挑战性开放问题的模型 · → daily
- 2026-06-14 ·
narrative · 技能文件编写能力评估 · → daily
- 2026-06-11 ·
narrative · 模型质量极高 · → daily
- 2026-06-09 ·
fact · price per M input tokens · → daily
- 2026-05-24 ·
narrative · 能力水平 · → daily
- 2026-05-23 ·
narrative · 是否为独立模型 · → daily
- 2026-05-21 ·
fact · 未受益于physics-intern harness · → daily
- 2026-05-11 ·
narrative · 被描述为 close to Stockfish level 7 · → daily
- 2026-05-05 ·
fact · can generate similar paper · → daily
- 2026-04-30 ·
narrative · 推理模式选择（正常/扩展） · → daily
- 2026-04-25 ·
fact · Mensa Norway score · → daily
- 2026-04-25 ·
fact · Mensa Norway average score · → daily
← 实体目录 · 系统日志