@thsottiaux I’ve felt that when working on AI models codex is never really able to get good results and sense hearing the claude stuff i’m worried Open AI also degrades performance on that type of work is that the case?
English
Jackson Stone
136 posts
Jackson Stone
@JacksonnCreator
DittoDub || 82B+ views generated– AI Dubbing for YouTube Creators. Trusted by MrBeast, IShowSpeed, John Nellis, MKBHD, SypherPK, Tucker Carlson, exc.
Utah Katılım Ocak 2019
16 Takip Edilen22 Takipçiler
Codex usage limits will be fully reset again in the next hour and we will credit one additional reset into your bank for your own usage over the next 24 hours.
We investigated reports that Codex usage was being consumed faster than expected. There wasn't one central issue, but a few smaller problems compounded for some users.
Here's what we found and changed:
- Actual usage: Auto-review had become more proactive, another change was triggering more subagent work, and background suggestions could run twice or retry too frequently after failures. We reverted the changes and fixed suggestion scheduling, duplicate generation, and retry behavior. This should reduce unnecessary background token consumption while preserving the work users explicitly request.
- Usage reporting: Auto-review was incorrectly appearing as GPT‑5.4 usage, and failed or rate-limited requests were still shown as turns. Auto-review now appears as its own category, and only successful requests count toward the turn graphs. Rate-limited requests were never charged, but they were being displayed incorrectly.
- Immediate relief: We reset usage limits while rolling out the fixes, then shipped hotfixes across the CLI, desktop app, and usage backend.
- What to expect: New usage data should be clearer and actual consumption should be lower. Historical charts may still show auto-review under GPT‑5.4 because older turn data was not relabeled. Features that intentionally perform more work; such as /goal, subagents, and higher reasoning levels will still naturally use more capacity.
All fixes are now deployed, and we've added more detailed monitoring so we can detect background-usage regressions sooner. We'll continue watching the results closely.
Thank you for building and doing all sorts of things with Codex.
Tibo@thsottiaux
Codex team is in a warroom on a Sunday combing through logs and checking whether there is anything that could lead to increased usage drains for some users. Taking it very seriously and won't rest until we get to the bottom of it.
English
@L0ONATUNES AI dubbing and English subtitles should stay separatein the YouTube workflow. Keep the original captionstrack live, then add dubbed audio as an alternate trackwhen the channel supports it.
English
One thing I keep seeing with YouTube dubbing: the upside is easier to understand once creators look at audience growth by language. The hard part is starting, then staying consistent enough for new viewers to trust the channel.
English
DittoDub is hiring a paid entry-level thumbnail editor intern. Looking for someone with a good eye for YouTube-style thumbnails, clean layouts, readable text, and fast iteration. DM us with examples if interested.
English
DittoDub is hiring a paid entry-level video editor intern. Good fit if you can cut short-form clips, move quickly, stay organized, and take feedback well. DM us with a quick intro and a couple edits/reels you've worked on.
English
We're hiring customer support at DittoDub. Looking for clear communicators who can help users, spot patterns, document issues, and keep customers feeling taken care of. DM us if you're interested, ideally with any support/SaaS experience.
English
DittoDub is hiring for lead generation. Looking for someone scrappy, organized, and good at finding/researching the right companies and contacts without spammy tactics. DM us with a quick note and relevant experience.
English
We're hiring a frontend/software engineer at DittoDub. If you like building fast, polished product UI and can ship clean React/TypeScript work without much hand-holding, DM us with a quick intro and a couple things you've built.
English
@AllHailDTiger @NoBody_here66 Auto translation feeling usually comes from flat timingand literal phrasing. For AI dubbing, review sentencebreaks and speaker emotion before final export.DittoDub focuses on this layer \(I\'m with DittoDub\).
English
@NoBody_here66 Motham Ai dubbing la undi direct Auto translation laga
Français
@soju967 @Single_Boy_AK Hindi and English versions work best when the dub keepstiming and intent close to the original scene, not justthe words. That\'s the quality bar we track at DittoDub\(I\'m with DittoDub\).
English
@Single_Boy_AK AI dubbing bro Hindi and English version best to watch 👍
English
@VINNUSAURUS That 70\% okay-ish mark is usually where timing,pronunciation checks, and mix consistency matter morethan raw translation. Small review passes can make AIdubbing feel much less patched together. \(I\'m withDittoDub\)
English
@VINNUSAURUS AI dubbing gets much better when the source audio isclean and speaker changes are obvious. Keeping a humanreview pass on names and slang matters too. \(I\'m withDittoDub\)
English
@dslicckest1 AI dubbing can get brittle fast on longer runs. For a1-hour file, split by speaker or scene first, then QAnames, timing, and emotional tone before stitching itback together. I\'m with DittoDub.
English
Jackson Stone retweetledi
