🕯️Ranol☆P 🦋 ranolp.dev

int mian() { return ;-; }

he/him. 성인.
언제나 차단 없음.
말같지도 않은 이유로 사람임을 포기하지 말자.

커서 프로 파딱 사면 내 북마크 편하게 정리할 수 있음.

[🕯️Ranol☆P 🦋 ranolp.dev @RanolP]

난노와 '더' 피자챗 파티! (2/28 예정)

이번엔 신림-봉천 인근에서 엽니다.
대학생 분들 방학 끝나기 전에 오세요.
연사자 환영합니다. 참가자 30명 제한입니다.

https://t.co/J3KzGikpBC

[🕯️Ranol☆P 🦋 ranolp.dev @RanolP]

퇵

[토키노 메리 @TokinoMary]

내일 휴방이라서 오늘 콜라 마셨어 행복하다
인형생 별 거 없따

[신초혜 @godCH_]

좃됨 ㅅㄱ

[화백 @matananoyo_uu]

헐 이거 14g 맞대!!!!
아니 어케 했냐???
삼각김밥으로 탄수 19g 지방 7g 단백질 14g 어케 했냐????????

[🕯️Ranol☆P 🦋 ranolp.dev @RanolP]

@NemuriaMiya @Basix1120 인간slop

[catnose @catnose99]

ついさっきBetter Authから
・Critical 1件
・High 2件
・Moderate 8件
の脆弱性修正のリリースが出てる

今v1.6以下を使ってる場合は1.7に上げる必要があるんだけど、破壊的変更もあるから注意

[굿부부 @Goob_old_days]

예체능 다해봄+입시학원 강사 n년으로 확실하게 말할 수 있는게 있습니다.

-악기=수학머리
-성악(노래)=영어머리

입니다. 이건 진짜임.... 물론 지가 안하면 말짱 꽝이니까 100퍼 성적나온다고는 말 못하는데 확실히 도움이 됨.... 괜히 옛날에 팝송으로 영어 배웠다고 하는게 아님

[AGF Korea @AGF_Korea]

@gemichinnomyeye 안녕하세요, 샤야입니다. AGF 2026의 RED STAGE는 12월 5일(토)과 6일(일) 이틀간 진행되며, 12월 4일(금)에는 RED STAGE가 운영되지 않는 점 안내드립니다.

[김 망멩미 @ReallyiAmGhost]

토스 CEO 화장실에서 휴지 쓰려고 하면 "잠깐! 페이스페이 등록하고 편하게 휴지 쓰세요. 30초면 돼요" 뜨고 그냥 닦을래요 누르면 한칸 나오고 "축하합니다 휴지 10칸 이용권에 당첨됐어요" 뜨고 받기 누르면 "페이스페이 등록하고 휴지 10칸 바로 받기" 나오게 해야됨

[Ryu @_ryu15_]

modal. com はまじで登録したほうがいいっすよ！

特にLoraとかで軽いんだけど自宅GPUでやるには重すぎるみたいな処理のときに、
modal使えば無料枠30$でA100とか借りて爆速で回せますからね

しかも一回限りの30$ じゃなくて毎月貰える(貯められない)ので限界地元民にはまじでオススメ

[Theo - t3.gg @theo]

This is a terrible compaction strategy that fundamentally doesn't understand how compaction and context management work.

Seems like a lot of people are confused so let's break this down.

1. Compaction isn't a filter
The role of compaction is to clean up history to keep the agent focused, not just deleting noise. It should be used sparingly when context gets too long, not constantly to keep context small.

2. Jev doesn't even know what it's deciding on!
Models use the context of the thread to decide what to keep or not keep in a summary. This implementation goes through on a "line-by-line" (per tool call) basis to decide what should be left or deleted.

Not only does this 32k token context model know very little of what happened before, but (in this implementation) it doesn't even know what the result of the tool call is!

Deleting these things randomly will keep the model from knowing what it's tried and dooms you to end up in "stupid loops" where the model keeps trying the same thing over and over.

3. You're giving up the reasoning entirely
Frontier models from OpenAI, Anthropic, XAI, and Google do not share reasoning traces over the API. They share encrypted payloads, which Jev cannot see (and often will drop).

Anthropic is even stricter with this, requiring you to preserve the entire history in order to get any of the reasoning data. As a result, using this in Claude Code guarantees the model will act way dumber.

4. Models are tuned on their compaction flows
For the last year, Frontier Labs have been including compaction and long runs as part of the training process. These models have learned ways to compact that are more effective than any rudimentary solution.

Fun fact: If you switch models in Codex and compaction is necessary, compaction will run on the model that was previously used in the thread.

5. Cache writes are more expensive than cache reads.
Cache writes are the biggest cost by far for agents. I often see cache write costs go over 60% of my total LLM spend in my personal use of Claude Code and Codex.

Cache writes are insanely expensive when data earlier in the history is changed (because the old cache is invalidated when things change at the top). Every history edit requires a cache rewrite for ANY data past the history edit.

If your history is "1,2,3,4,5,6" and you delete "2", you have to rewrite "3,4,5,6". This is more expensive than leaving "2" in the history.

Good news. Since we're already killing all of the reasoning tokens by doing this stupid compaction strategy, the rewrite cost won't actually be that high because the model is missing so much data! 🙃🙃

6. The implementation is hot garbage.
> "Whatever is not kept is deleted permanently, but the assistant can always re-run a tool or re-read a file."

Good luck with that one.

To be clear: this is a cool experiment and I find it genuinely interesting. That said, if you think this style of bs filtering on a probability threshold is actually a compaction strategy, I highly recommend you just use the defaults in tools like Claude Code and Codex.

You're much less likely to hurt yourself that way.

[준희 @bmtxb]

이래서 마케팅관련 직무는 지원자의 학점보다
포트폴리오랑 개인역량을 더 많이 보는구나 ...

[Basix @Basix1120]

안녕 트친들 나를 믿고 https://t.co/PFLN0Od0ux 를 설치해주지 않을래

[菅原道真の化身(アメリカから福岡に左遷) @J8mtk]

³√２の無理数の証明
³√２=p/q
2=(p/q)^3
2p^3=q^3
p^3+p^3=q^3
フェルマーの定理最終定理より流石にこれは存在しない　◾️

[🍊Austin Hardwicke 🍊chompotron.bsky.social @chompotron]

Exact same amount of frames, completely different feel. Seriously a masterclass

[Basix @Basix1120]

jj를보고 줴줴이야라고 읽으면안돼요

[ParrotDash 🦜 @ParrotDash]

We got Celeste 2 Player maps now

[@슈-크림《환생맛》 @ashu_kurimu]

@Qudckarhk 적십자 마크 사용하는 방법 : 콜라보를 한다
이거 외에는 없는듯

[병참과의 총기봇 @Qudckarhk]

아무튼.... 적십자사 입장에서는 진짜 구호가 필요한 상황에서 사람들이 조금이라도 적십자사의 구호물품이나 표장을 신뢰하지 못하는 상황을 만드는 걸 방지하기 위해서 하는 조치들이니 가급적이면 다들 따라주시기 바랍니다.

그리고 이게 더 멋진 디자인을 만드는 계기가 될 수도 있어요.

[채고로강하다 @chaegang0326]

#solved_art
헤어스타일 체인지