Wes Roth · 2026-07-09 · notable
Wes Roth: 'GPT-5.6 is here (INSANE)'
Wes Roth walks through OpenAI's GPT-5.6 general-availability launch — Sol, Terra, and Luna — and the new max reasoning effort and ultra mode subagents, uploaded within hours of the announcement.
Wes Roth reacts to OpenAI's GPT-5.6 launch, hours after Sol, Terra, and Luna go public.
What is it?
Wes Roth uploaded 'GPT-5.6 is here (INSANE)' on July 9, breaking down OpenAI's general-availability launch of GPT-5.6 Sol, Terra, and Luna within hours of the announcement. The video walks through Sol as the flagship reasoning tier, Terra as the balanced everyday model, and Luna as the low-latency variant now live across ChatGPT, Codex, ChatGPT Work, and the API.
How does it work?
Wes Roth stitches OpenAI's blog post and X thread with live examples of the new max reasoning effort setting — which gives Sol more inference time on hard tasks — and ultra mode, which spins up subagents so a single Sol request can parallelise steps. Roth pairs the walkthrough with commentary on what the tiered pricing means for people already paying for GPT-5.5.
Why does it matter?
Roth's channel is one of the fastest AI news feeds on YouTube, and this video is the first mainstream explainer of the GPT-5.6 rollout for viewers who do not read OpenAI blog posts. The reaction format is how most casual users decide whether to upgrade, so the framing here shapes how GPT-5.6 lands with the wider dev audience today.
Who is it for?
Developers and casual ChatGPT users tracking OpenAI's tiered model rollouts.