ModelsAug 17, 2026 · 06:405 sources
Codex Caps Subscription Context at 360K Tokens Defying Official 1M Token Guide
From huggingnews.com
Codex subscription users reported a practical context limit of 360,000 to 371,000 tokens for the GPT 5.6 Sol model, despite documentation for a 1,050,000 token window. Using expanded context increases the drain on usage limits, with tokens beyond the default boundary reportedly counting at twice the standard rate.
Read the sourceSign in to suggest
Sources
+ 4 more
- SUPPORT@teknium“actual server-side max for codex subscriptions is just 360~K”x.com
- SUPPORT@teknium“371k is the actual absolute max”x.com
- SUPPORT@reach_vb“Tokens beyond the default context window count at 2x against your usage limits”x.com
- SUPPORT@kimmonismus“anything over 272k would consume usage limits at 2x the rate”x.com
One ranked edition, every morning
The stories that matter today. Free, cancel anytime.
Telegram @aidr_today