We’re rolling out a change to increase cache hit rate on Neon by around 4% on Antigravity by massive increasing the cache pool size. This will increase what you can get done in quota by at least 20%. Will issue a reset after it’s done since it burned quota faster!
AI TrendsReset monitor
Follow Varun at Google for Antigravity usage resets, timing, eligibility and original announcements.
Quota reset is within its estimated window
Within estimated window
Estimated 11/21 02:38–11/21 14:38 UTC+8
This timing is an estimate, not confirmation. Check your actual usage in Antigravity.
- Usage resets · 90 days
- 6
- Median reset interval
- 12.5 days
- Last usage reset
- 2026-11-18
Reset calendar
Choose any date to see records and source posts.
Browse 15 recorded dates
This month · 0 resets · 0 pending · 0 unconfirmed
Recent source posts30 public records
Varun@_mohansolo@OrganicGPT We’re working on it!
Replies and quoted context (1)
Reply to @OrganicGPTCache hit rate >80% in 2026 is pretty standard. I Get +99% on DeepSeek Harness and V4.1 Flash. Unfortunately, Gemini 4 series has abysmal hit rate https://t.co/jfP2swCw8i
Varun@_mohansolo@aicodeking Should be fixed soon! https://t.co/3vRHpOXVvr
Replies and quoted context (3)
Reply to @aicodekingI investigated this a bit and it seems that the cache TTL for Antigravity is somewhere near 5mins. It keeps hitting 90-98% cache until ~5 mins of idle time and then it goes downhill to 9%. I haven't noticed this because of long-horizon tasks but it does seem to happen. The 15-30s or 30-60s or 1-2mins claim is not correct for me. It only ever happens after 5mins. But, this shouldn't happen as TTL on the Gemini API sits at about 1hr and Claude, Codex holds about 85% throughout that period as well.
Quoting @_mohansoloWe’re rolling out a change to increase cache hit rate on Neon by around 4% on Antigravity by massive increasing the cache pool size. This will increase what you can get done in quota by at least 20%. Will issue a reset after it’s done since it burned quota faster!
Quoting @theoOnce the honeymoon phase wears off a lil, I’ll go back to harassing Google into fixing these things. Neon is the first Gemini model that’s good enough to actually deserve a functioning platform.
Varun@_mohansoloMore info: The hit rate was particularly bad for workloads where there was more time between model invocatikns (cache was evicted). This should fix that and they should see a much higher improvement. We will continue to improve this metric globally, sorry for the inconvenience!
Replies and quoted context (1)
Reply to @_mohansoloWe’re rolling out a change to increase cache hit rate on Neon by around 4% on Antigravity by massive increasing the cache pool size. This will increase what you can get done in quota by at least 20%. Will issue a reset after it’s done since it burned quota faster!
Varun@_mohansoloAnnouncementWe’re rolling out a change to increase cache hit rate on Neon by around 4% on Antigravity by massive increasing the cache pool size. This will increase what you can get done in quota by at least 20%. Will issue a reset after it’s done since it burned quota faster!
Varun@_mohansolo@shneural DM? Would love to dig since this is quite unexpected
Replies and quoted context (1)
Reply to @shneuralGemini 4 Neon is burning your Antigravity limits ~3x faster. prompt caching is broken (9% hit after 1m) Opus 5.5: ~95% (1h) GPT in Codex: ~97% (24h) sucks cuz Neon is actually insane for agents. Google i know infra is rough rn, but pls fix (put Neon on it lol + reset limits) https://t.co/sBqP9RprUt
Varun@_mohansolo@aicodeking Interesting, it should not be polling this much. The model should set a fairly reasonable length timer and use that.
Replies and quoted context (1)
Reply to @aicodekingGemini Neon is really good.. Just one feedback... This is probably a harness related one.. A long command can only stay in the foreground for 10 seconds. After that it goes to the background. To wait for it, manage_task status needs WaitDurationSeconds. 0 means "tell me now." 300 means "wait until it finishes." That field is required. If you leave it out, agy does not say the call is wrong. It answers "still running" right away, same as 0. The agent then asks again on every turn. I hit this on a 30 minute render. 195 status checks in 15 minutes. The render was still going at about 48% when the session died. There were 657 status checks in that session. Please reject a status check that has no wait time, or treat a missing wait as 300, not 0. Neon does this more often when using via ACP. @_mohansolo @OfficialLoganK please fix this if possible.
Varun@_mohansolo@wolfie_ Strange, can you share your error ID?
Replies and quoted context (2)
Reply to @wolfie_@_mohansolo Neon High
Reply to @_mohansolo@wolfie_ Hmm, what model are you using?
Varun@_mohansolo@wolfie_ Hmm, what model are you using?
Replies and quoted context (1)
Reply to @wolfie_antigravity is unusable, can't go more than a few minutes without hitting "Agent execution terminated due to error"
Varun@_mohansolo@BenjaminDEKR What are you seeing in terms of time outs @shengzheyao
Replies and quoted context (1)
Reply to @BenjaminDEKRGoogle: nice models, guys. Now please fix your CLI harness so it stops timing out and is actually usable! 🙏
Varun@_mohansoloAnnouncementOne more thing… In honor of Antigravity turning 1, we’re resetting quota for everyone across all models!
Varun@_mohansoloAnnouncementTo mark a week of relentless shipping, we’re resetting everyone’s Gemini quota on Antigravity! Go build something amazing with swarms 🚀
Varun@_mohansolo@HarshithLucky3 Come on, can we just enjoy the model today without a reset 😀
Replies and quoted context (3)
Reply to @HarshithLucky3@_mohansolo give us reset varun https://t.co/qZqxhTgvsR
Reply to @_mohansoloGemini 4 Carbon is now available for Ultra users in Antigravity (and yes, we will quickly expand access to more users). We are also launching /swarm as a native experience to leverage multi agents to complete complex tasks. Swarms have been a step function improvement for capability internally, especially with Carbon.
Quoting @HarshithLucky3Gemini 4 Carbon is showing in my Ultra sub but i dont have limits @_mohansolo give us a reset please https://t.co/d6NnXI3vzh
How is the timing estimated?Sources and rules
Explicit times, relative hours and deadlines are interpreted in America/Los_Angeles and converted to Beijing time, with a small allowance for rollout delay.
A date-only announcement uses the whole Pacific calendar day. Without a stated time, the initial estimate is the following twelve hours; there is no assumed daily reset schedule.
After an estimated window ends, records await confirmation for six hours, then show likely completion. An estimate never becomes an officially confirmed fact.
The configured scan interval is ten minutes. Source announcements, not this page, determine eligibility; personal usage is only available in the product.
Records begin 2026-05-21Needs attention · Last checked 11/21 06:34AI News · Not affiliated with Google
- Last attempted read
- 11/21 06:34
- Last posts collected
- 11/21 05:54
- Last complete check
- 11/21 06:34
A recent source reported an outage.View source