A quick company-evaluation test says Fable found positioning issues, while Sol just added documentation theater
jdjohnson · x · 2026-07-27
The author says they routinely test models by asking them to evaluate their company and recommend three moves that would improve it immediately.
- Fable was the first model that surfaced anything useful, mainly around positioning.
- Sol was judged useless: it mostly generated extra documentation no one needed, and when challenged it often admitted the suggestion would not actually help.
The post is essentially a quick comparison of how different models behave on a practical business-analysis task.
More from Models
- llama.cpp merges GLM-5.2-Vision support for local multimodal inference — QuixiAI · 2026-07-27
- Elon Musk calls Grok 4.5 a solid workhorse after Tim Sweeney’s praise — elonmusk · 2026-07-27
- Open models rose from 10% to 30% of tokens in a year, Together says cost is the reason — togethercompute · 2026-07-27
- Hyperagent pits Opus 5 against GPT-5.6 Sol on real browser-agent tasks and the cheaper model holds up — PrajwalTomar_ · 2026-07-27
- Ethan Mollick says Opus 5 and Codex voice mode forced a same-day model guide update — emollick · 2026-07-27
- Alibaba should ship multiple Qwen sizes, says researcher focused on interpretability — traviscline · 2026-07-27