@julien_c It's very strange spring... First it was very cold ~15C until early May then it's suddenly +32C 🫠here in Chartreuse 🌲
English
Mikhail Shevtsov
386 posts
Mikhail Shevtsov
@mesouug
Entrepreneur KraftStream Founder 🚀 https://t.co/9ZTlT4PWNx Perfectionist
France Katılım Haziran 2017
77 Takip Edilen24 Takipçiler
Who else feels like they lose ~20 IQ when it’s hot outside?
English
@Adam_Iverson @ClementDelangue Try following settings. See screenshot. It works well on my RTX3090 in Docker via presets ini config
English
@ClementDelangue For some reason, I couldn't get it to work on RTX 3090, but it works quite well with the M1 Max MacBook Pro. Both are using the latest version of LM Studio. Not sure what's happening.
English
@ASUS Thank you very much @digitalix for your inspirational video ❤️I was quite nervous spending 3000 euros on unknown device 😅
English
@Slant3D If I didn't know how it feels in hand 👋 I would say - RIGHT one! Buuut there is always a butt. The LEFT one is much more satisfying to touch. So I would choose the LEFT one
English
@Slant3D Is this an option when ordering part... Like choose final finish?
English
Do you think your friend would appreciate this? 🤔🐙
Check out our full tutorial for more 3D model editting tips: youtube.com/watch?v=nPC6pd…
3D models in the video:
nexprint.com/en/models/G171…
nexprint.com/en/models/G013…
YouTube
English
@LowLevelTweets Some things needs to be dumb simple. I don't need AI in my fridge. I need more real estate. Today I discovered that I can remove freezer tray and I can stuff it with 3x more food than with tray installed 😮
English
@adamdotdev Lol... I found $500 on 5 year old bitcoin account 😃 If you look closer it's actually a very cool technology as long as it's not banned by the officials 🙃
English
@UnslothAI @Alibaba_Qwen Would be could if you include hardware details on your graphs! 😃 Also interesting to learn VRAM usage vs "spec-draft-n-max" parameter 🤔
English
Qwen3.6 now runs 2x faster with MTP GGUFs! Run locally on just 18GB RAM. ⚡️
MTP enables Qwen3.6 to generate ~1.4–2.2× faster with no accuracy change.
Qwen3.6-27B MTP runs at 160 tokens/s. 35B-A3B reaches 240 t/s.
GGUFs: huggingface.co/unsloth/Qwen3.…
Guide: #mtp-guide" target="_blank" rel="nofollow noopener">unsloth.ai/docs/models/qw…
English
Yay, people starring my repo. Hope this new bench is handy for coders.
English
@digitalix Finally! I never thought this would ever end. These artists keep surviving every damn time 🫠
English
Now it’s REALLY OVER!
fabian@fabianstelzer
Anthropic launches Claude Monet Painters are cooked
English
@ryanvogel The best known strategy - is quarantine private mirror - the only downside is cumbersome to setup and maintain - but absolutely must have for popular project - see @videolan - they compile GCC first and then VLC on air-gapped machine for the release
English
This stuff is getting more and more common every day, is there any guide or tutorial that shows how to properly secure a github account/repo/org/etc?
Socket@SocketSecurity
🚨 BREAKING: 84 TanStack npm packages were compromised in an ongoing Mini Shai-Hulud supply chain attack, adding suspected CI credential-stealing malware. Socket flagged every malicious version within six minutes of publication. This is a developing story.
English
@ryanvogel Yeah... it's really too cumbersome - I don't like it. Not sure if this possible... But maybe use opencode over ssh like X11 session path-through or how rsync does it. I think it would be much more reliable solution rather than launching opencode on remote machine directly
English