Many users praised OpenAI's transparent updates and quick reverts on GPT-5.6 Sol context limits and usage bugs for building trust, while others criticized the initial release of buggy features causing excessive Codex usage.
Based on 74 visible X reactions from 201 accounts.
Ask a question below.
Published answers will appear here.
@MatthewBerman My experience with codex and ChatGPT work has been nothing short of awesome.
@theo Thank you for always being there for your customers. I'm rooting for you.
@scaling01 This is why you are the Usul Thank you so much
There's a few issues with Codex that stacked really badly, causing people's usage to get nuked. Crazy how well they all lined up to be as rough as possible Issue 1 (shown here): gpt-5.6 costs 2x more after 272k tokens. Codex upped their limit to 372k tokens, meaning long threads were billed 2x higher Issue 2: "Ultra" subagents were also spawned with Ultra, causing insane nested subagent usage that burned a TON of tokens. Apparently intended, but (hopefully) being toned down in system prompt Issue 3: Sol and Terra use the "v2" subagent layer in Codex that is still early/unfinished/disabled by default. v1 spawned subagents with a fresh history, v2 copies the entire long context. When combined, you end up with: 1. A ton of reasoning tokens filling context windows, triggering 2x billing 2. Subagents spawning with that filled context window, instantly billed at 2x 3. Ultra spawning too many of these high reasoning full-window subagents If you had fast mode on when this happened, you got hit with an ADDITIONAL 2.5x No wonder we were burning so hard...
(1) is not correct, it is not due to 2X charging after 272k context, we don't charge for longer context on the subscription for GPT 5.6 Sol as we control all the settings. It is due to something else, which I will attempt to explain below. The overall trajectory length, which is the total some of all context windows across compactions, changes little based on the reasoning effort. Similarly the quality of the overall output is similar across context lengths above 272k. The benefits of higher context lengths are mostly overall speed (as you don't wait for compaction), ability to deal with humongously large input and potentially cost if the system is well tuned and you hit your cache perfectly. The actual reason is the what you can see depicted in the chart below, which is the difference in the orange line and the blue line. It is caused by overall cost of cache reads going up with the size of the context being shuffled back and forth between toolcalls. The sweet spot in terms of cost is therefore not necessarily to use the maximum possible context length. What we're working on is tuning the system differently so that we can go back higher without it resulting in higher usage being charged. Hope this clarifies a bit.
The temporary rollback to 272k tokens prevents unintended usage charges.
@MatthewBerman My experience with codex and ChatGPT work has been nothing short of awesome.
@theo Thank you for always being there for your customers. I'm rooting for you.
@scaling01 This is why you are the Usul Thank you so much
There's a few issues with Codex that stacked really badly, causing people's usage to get nuked. Crazy how well they all lined up to be as rough as possible Issue 1 (shown here): gpt-5.6 costs 2x more after 272k tokens. Codex upped their limit to 372k tokens, meaning long threads were billed 2x higher Issue 2: "Ultra" subagents were also spawned with Ultra, causing insane nested subagent usage that burned a TON of tokens. Apparently intended, but (hopefully) being toned down in system prompt Issue 3: Sol and Terra use the "v2" subagent layer in Codex that is still early/unfinished/disabled by default. v1 spawned subagents with a fresh history, v2 copies the entire long context. When combined, you end up with: 1. A ton of reasoning tokens filling context windows, triggering 2x billing 2. Subagents spawning with that filled context window, instantly billed at 2x 3. Ultra spawning too many of these high reasoning full-window subagents If you had fast mode on when this happened, you got hit with an ADDITIONAL 2.5x No wonder we were burning so hard...
(1) is not correct, it is not due to 2X charging after 272k context, we don't charge for longer context on the subscription for GPT 5.6 Sol as we control all the settings. It is due to something else, which I will attempt to explain below. The overall trajectory length, which is the total some of all context windows across compactions, changes little based on the reasoning effort. Similarly the quality of the overall output is similar across context lengths above 272k. The benefits of higher context lengths are mostly overall speed (as you don't wait for compaction), ability to deal with humongously large input and potentially cost if the system is well tuned and you hit your cache perfectly. The actual reason is the what you can see depicted in the chart below, which is the difference in the orange line and the blue line. It is caused by overall cost of cache reads going up with the size of the context being shuffled back and forth between toolcalls. The sweet spot in terms of cost is therefore not necessarily to use the maximum possible context length. What we're working on is tuning the system differently so that we can go back higher without it resulting in higher usage being charged. Hope this clarifies a bit.
Unprecedented levels of transparency out of OpenAI, really cool to see. They definitely made mistakes but they’re addressing them with full transparency https://twitter.com/thsottiaux/status/2076495156757577895
@thsottiaux Ah this makes sense - thanks for clarification! I can see how this would add up pretty bad, surprised how much it seems to though Is the plan to make cache reads cheaper? 👀👀
Tibo has clarified - point 1 is not correct. The tokens over 272k are not at 2x, it just compounds brutally after a bunch of compactions https://x.com/thsottiaux/status/2076543065045795309
My obsessive community noters have gotten EXCEPTIONALLY bad at reading as of late. https://x.com/theo/status/2076522097460027821/photo/1
Also before I forget - 100% of the advice in my "save your limits" article still applies:) https://x.com/theo/status/2076078865060151465
https://x.com/i/status/2076495156757577895
Many users praised OpenAI's transparent updates and quick reverts on GPT-5.6 Sol context limits and usage bugs for building trust, while others criticized the initial release of buggy features causing excessive Codex usage.
Based on 74 visible X reactions from 201 accounts.
Ask a question below.
Published answers will appear here.
Unprecedented levels of transparency out of OpenAI, really cool to see. They definitely made mistakes but they’re addressing them with full transparency https://twitter.com/thsottiaux/status/2076495156757577895
@thsottiaux Ah this makes sense - thanks for clarification! I can see how this would add up pretty bad, surprised how much it seems to though Is the plan to make cache reads cheaper? 👀👀
Tibo has clarified - point 1 is not correct. The tokens over 272k are not at 2x, it just compounds brutally after a bunch of compactions https://x.com/thsottiaux/status/2076543065045795309
My obsessive community noters have gotten EXCEPTIONALLY bad at reading as of late. https://x.com/theo/status/2076522097460027821/photo/1
Also before I forget - 100% of the advice in my "save your limits" article still applies:) https://x.com/theo/status/2076078865060151465