AI Weekly Malaysia

Back to items Summaries

GPT‑6.2 Astra and Sol

ID
47511
Status
summarized
Published
18 Nov 2026, 1:46 AM
Fetched
18 Nov 2026, 3:54 AM
Provider
Hacker News
Category
dev-community
Original URL
https://openai.com/index/introducing-gpt-6-2-astra-sol/
Source URL
https://hnrss.org/best

Summary

Score
7.0
Created
18 Nov 2026, 3:54 AM
Tags
Audience
developersai_ml_learnersai_agent_userssaas_foundersvibe_coders

What happened

OpenAI announced GPT-6.2 Astra (frontier model for hard, long-horizon work) and GPT-6.2 Sol (cheaper everyday/agentic model), claiming both need fewer output tokens per task. Astra's API pricing drops 20% versus GPT-6 Astra to $8 per million input tokens, $40 per million output, and $0.40 per million cached input — a 60% cached-input discount — while Sol keeps GPT-6.1 Sol's prices. OpenAI reports Astra beating Claude Fable 5.2 by 5.5 points on Agents' Last Exam at 5% of the cost, and Sol landing within 0.5 points of Astra on Cognition's FrontierCode 2.0 at roughly one-fifth the cost; the HN thread drew 623 points and 359 comments.

Why it matters

The cached-input price ($0.40/M, 60% off) is the number to check against your own agent loops: retrieval-heavy or repeated-system-prompt workloads get materially cheaper, while output at $40/M stays the cost driver, so token-efficient prompting and output caps still matter more than the headline 20% cut. Treat the Agents' Last Exam and FrontierCode 2.0 gaps as vendor-reported until you rerun your own eval set — the Fable 5.2 comparison already excludes nine tasks that used a fallback model, which is a concrete reason to distrust the ratio.

Discussion angle

Before switching, run your own cost-per-completed-task math on one real agent workflow — cached input at $0.40/M versus output at $40/M means the split of your tokens, not the 20% headline, decides whether Astra or Sol is actually cheaper for you.

Top