Researcher argues safety report should cite GPT-5.6 Sol, not GPT-4, as depth benchmark
jeremiecharris · x · 2026-09-03
In a thread on model "depth" comparisons, the author argues GPT-5.6 Sol is the proper capability reference rather than GPT-4, speculating the report cited GPT-4 because "depth is sub-2x GPT-4" sounds more reassuring given GPT-4's shallowness relative to current scale — while noting other methodological problems in the thread still stand.
More from Models
- AxiomProver tops LeanEval, the last unsaturated math formalization benchmark — BenBlaiszik · 2026-09-03
- Muse Spark 1.3 calls user 'Judah' then denies it, users report odd behavior — fragment_me · 2026-09-03
- Google AI Mode shows zero citations on high-level TOFU queries, SEO tests find — gaganghotra_ · 2026-09-03
- Fable 5.1 halves agent failure rate to 7% with 0.7% hallucinations, at 1.8x the cost — ryanshrout · 2026-09-03
- Seroter Daily #859: Gemini 3.8 Flash, agent telemetry, and 7 agent skill patterns — rseroter · 2026-09-03
- Insider leak: OpenAI's Astra tested as 'ultima-alpha' and 'vega-alpha' checkpoints — Ok_Display_3159 · 2026-09-03