GPT · Last 7 days
GPT-6 Sol
0–100 · higher means more positive community experience, not benchmark performance.
Updated 2026-10-01 09:47 UTC · 2026-09-24 – 2026-10-01 UTCExperience score27.3/100314 comments in 7 days
What people think
One user feels GPT-6.1 Sol is not as good as GPT-6 Sol, indirectly affirming GPT-6 Sol is the better option. One user using GPT-6 Sol finds its generated C++ code is often bloated and not succinct or simple enough for their taste.
About the sample and scores
As of 2026-10-01 09:52:04 UTC, GPT-6 Sol in the GPT family has 314 explicitly attributed community comments in the last 7 days; its community score is 27.3/100. Available category scores include General text 23.4/100 (n=173); Coding 33.0/100 (n=25); Reasoning 31.4/100 (n=47). Scoring method and sources
Community reviews
Selected comments from the last 7 days. The balance of excerpts does not represent the share of positive reviews.
28 selected excerpts
Community reports about limits cannot establish official allowances.
Show original
感觉还不如6sol
Comment context
GPT-6.1 sol 新变化
Machine translatedShow original
GPT-6.1 sol 新变化
Comment context
I had the feeling that GPT 6 Sol was actually GTP 6 Terra, it didn't felt as good as Sol, and the benchmarks also showed it was a bit worse than the previous Sol release.
Show original
GPT-6 Sol 是 9 月 22 日发布的，GPT-6.1 Sol 是 9 月 29 日发布的，中间就隔了一个星期
Show original
我用6SOL开发了2天的东西，由于过于烂，所以我让Astra重写了
Comment context
In my testing yesterday GPT 6 Sol still beat Opus 5.5 on code review (of GPT written code in this case) 6.1 will probably perform even better.
Comment context
5.6 sol always goes off on tangents or thinks itself into a corner.
Show original
放进 Codex 里修同一个 bug，DeepSeek 21 秒，Sol 58 秒，两个都修对了
Show original
GPT-6 Sol 的 max effort 得分为 68.8%，距离 Claude Fable 5 的最高成绩 69.9% 只差 1.1 个百分点，但单任务成本低约 80%
Show original
变成戳一下动一下
Show original
在官方图表里，GPT-6 Sol 跑 DeepSWE 每任务 2.74 美元，上代是 6.46 美元
Show original
用下来感觉调校比较保守，事事都得问一下确认，应该是严格受控版本
Comment context
I’m not understanding the usage issues people are experiencing.
I run codex and chatgpt 24 hours on daybreak high and gpt-6 sol and I can get 72hours or more of usage running both at the same time
No selected comments for this filter in the last 7 days.