OpenAI28-day sprint Day7：GPT-6.1 Sol default context expanded to 500000, with no long-context surcharge for subscribers
BIBIBI
AT A GLANCE
According to reports,GPT-6.1 Sol increased the default context to 500000 token, with no long-context surcharge for subscribers.
Article
OpenAI28 The final sprint Day7: Sol's default context expanded to 500000, with no long-context surcharge for subscribers
Beating AI news flash,OpenAI product and platform lead Thibault Sottiaux announced 28-day product sprint, Day 7。GPT-6.1 Sol in Codex and ChatGPT Work, the default context increased from 272000 token to 500000, an increase of approximately 80%. The larger window can reduce the number of times context is automatically compressed during long tasks, allowing the model to retain more details from earlier conversations and tasks.
Sottiaux said that this default value had undergone dedicated testing. Users can still adjust it higher or lower themselves, with support for up to 1000000 token. The model previously supported manually setting a larger context; what has increased this time is the default value.
At the same time,OpenAI no longer charges GPT-6.1 Sol using ChatGPT subscription quotas a long-context surcharge.
Original link https://m.theblockbeats.info/flash/371276
Key points
01
reported that GPT-6.1 Sol in Codex the default context in and ChatGPT Work increased from 272000 token to 500000 token, an increase of approximately 80%。
02
reported that a larger context window can reduce the number of times context is automatically compressed during long tasks.
03
reported that users can still adjust the context higher or lower, with support for up to 1000000 token。
04
reported that OpenAI no longer charges those using ChatGPT subscription quotas GPT-6.1 Sol a long-context surcharge.
05
attributed the above information to OpenAI product and platform lead Thibault Sottiaux.
AI-assisted interpretation
The following is analysis, separate from reported facts. Verify important claims independently.
A context window can be understood as the amount of conversation and task information the model can retain during a single task. After the default capacity increases from 272000 to 500000, long tasks may require less compression of earlier information; the report also says that eligible ChatGPT subscription quotas are no longer charged a long-context surcharge.
Why it matters to readers
If these adjustments apply to the relevant users, long tasks may retain more information, and the additional cost of using long context may also decrease.
For beginners, this means that when processing long documents, lengthy conversations, or complex tasks, the model may find it easier to remember earlier content, and the subscription usage method may be simpler.
Risks and unknowns
The evidence comes from a media news flash and does not provide OpenAI an official original announcement.
The evidence alone cannot confirm whether this adjustment has been made available to all users.
It cannot be confirmed which ChatGPT subscription tiers, regions, or users are covered by the cancellation of the long-context surcharge.
It cannot be confirmed whether Codex and ChatGPT Work are actually available to the relevant users.
Related Developments
Loading event timeline…
Related concepts
token
This term is not in the glossary yet. Browse related concepts in the glossary.
