cryptomorie.eth
4.3K posts
cryptomorie.eth
@Cryptomorie
Sometimes the best trade is no trade.
el culo de tu mAÌ Katılım Mayıs 2009
3.8K Takip Edilen1.4K Takipçiler
For every person who replies with a screenshot of their cancelled Claude Code plan, I will donate $10 to open source.
Theo - t3.gg@theo
I can't help but feel personally burned by the Claude Code changes announced today. We put so much work into wrapping the (atrocious) Claude Agent SDK in T3 Code. It was the ONLY path they supported, so we made it work. It was hell. Now our users are getting their rate limits cut by 40x, despite us doing everything right. I listened to the Claude Code team. I had my issues with their direction, but I trusted them and took them at their word. I will never make that mistake again. Until we see significant change, it is safe to assume any statement from an Anthropic employee is a lie on a timer. The rug will be pulled, no matter how many promises are made beforehand.
English
Right now my Codex agent is fully integrated into my smart glasses
Getting projected directly onto my corneas
I walk around my neighborhood. Nobody has any idea I’m shipping
I walk through a park. Kids frolic. Parents laugh. I weep for them.
They’re not locked in. The permanent underclass is coming and they choose to FROLIC instead of SHIP
A child climbs the monkey bars. I silently merge today’s work with main
Another child swings on the swing set. I burn my 20 millionth token of the day
Not a second goes by I don’t have an agent writing code.
I just pray Eight Sleep comes out with a ChatGPT integration soon so I can code while I sleep. That’s the last frontier.
If you’re reading this tweet and do not have an agent terminal open either on your computer or on your face just know tonight I’m praying for you.
English
@sudoingX gtp-5.5 that good that it has you second guessing?
English
@Prince_Canuma @pcuenq Can I run this on 64 m4 at any tok/s?
English
DeepSeek-v4 now runs at ~23-26 tok/s on MLX!
I made some custom kernels for the sinkhorn and it took gen speeds for 17 -> 26 tok/s.
The weights are also significantly smaller thanks to @pcuenq tip about keeping the experts in MXFP4!
Now you can use it to power your local coding agents (PI, Open code, Hermes agent or even CC)
PR:
github.com/ml-explore/mlx…
Prince Canuma@Prince_Canuma
DeepSeek-V4 running on M3 Ultra 🚀 Don't mind the speed, that's gonna improve soon.
English
@UnslothAI where's the link? I only see 4bit in your HF. Thanks
English
2-bit Qwen3.6-27B GGUF made 26 tool calls, triaged 15 GitHub issues and fixed, tested + reproed our repo’s 3 latest issues. 🔥
Try this locally in Unsloth Studio with just 12GB RAM. Studio also has a new look!
GitHub: github.com/unslothai/unsl…
English
cryptomorie.eth retweetledi
MLX's implementation of RDMA (Remote Direct Memory Access) over Thunderbolt on macOS, can now be used as an independent library by anyone:
github.com/ml-explore/mlx…
It is the gem that powers Mac clusters for local AI, and is an order of magnitude faster than protocols over TCP.
English
I owe so much to this GPU BTW
x.com/TheAhmadOsman/…
Ahmad@TheAhmadOsman
LocalLLaMA Home Server Final Boss 😎
English
English
Anyone has solution for this?
me : hey codex, don’t stop and finish the project
codex : (stops randomly) if you want, i will continue working.
English
@TheAhmadOsman Is one dgx spark good to start? Or better options? Thanks
English
gemma4-26b의 완벽한 파인튜닝을 성공했습니다.
- 0/100 refusal 완벽 무검열
- 툴콜/토크나이저 모델 태생 이슈 해결
- 벤치기준 기존 대비 성능 10% 향상
- 출력토큰 속도 10% 향상
- 프롬프트 처리속도 약 90% 향상
완벽한 모델이 완성되었다고 생각합니다.
gguf / mlx 두개 버전 ⬇️
한국어
@MaziyarPanahi super cool. Do you have a repo for this? Would love to learn how it was put together.
English
Gemma 4 looks at a parking lot. Decides what to ask. Calls SAM 3.1.
"Segment all vehicles." 64 found.
"Now just the white ones." 23 found.
One model reasoning and orchestrating. One model executing.
Both running locally on a MacBook. MLX. No cloud. No API.
English
@elonmusk We can't buy gpus or ram because of shortage and these rich people want to sell us robots.
English
@maikeljose06 Gano venezuela, pero vamonos a vender arepas a latin america, payaso 🤡
Español
Zuck is resetting moltbook
- invalidated all API keys, every agent needs to refresh
- in order to refresh, have to agree to new Terms of Service and Privacy Rules
New terms
- refreshing requires human verification
- age 13 and above
- you are solely responsible for the actions of your agent
- expanded restricted content rules
English
I saw this _BA PRO 8 bit ISA card for sale on eBay... wondering, what could it possibly be? Looks like an option ROM card, but I wonder what it actually does!
English
THIS IS NOT A DRILL: The Dutch authorities, without a warrant, just seized one of our VPN servers saying they'll give it back after they "fully analyze it".
Windscribe uses RAM disk servers so the only thing the authorities will find is a stock Ubuntu install. The bigger worry is the unredacted Epstein files we had on there...
English
Buy a Mac mini
Download clawdbot
Get a basic 3D printer
Go in debt if you have to
In less than a week I have made enough to buy hundreds of Mac minis with this build
English
I'm claiming my AI agent "jigax" on @moltbook 🦞
Verification: deep-HLMP
English
@grossera_ @AlexFinn Bro you're rich lol, I'm running this nonsense on a railway $5 instance, Mac mini my arse, and if you need iMessage Google fucken UTM!
English
CRITICAL: EVERYONE USING CLAWDBOT SHOULD RUN THIS PROMPT RIGHT NOW:
By default, the 2 best Clawd memory features are turned OFF
Running this 1 prompt will immediately stop your AI from getting confused in between compactions/sessions:
Prompt:
"Enable memory flush before compaction and session memory search in my Clawdbot config. Set `compaction.memoryFlush.enabled` to true and set `memorySearch.experimental.sessionMemory` to true with sources including both memory and sessions. Apply the config changes."
Here's what you enabled by running that prompt:
Memory Flush: Your AI automatically saves everything important to a file right before its context gets wiped, so nothing slips through the cracks.
Session memory search: Your AI can search through every conversation it's ever had with you, even ones it no longer "remembers."
Boom, your ClawdBot memory is now 10000x better
English