Andrew (@andrlfv) header
Andrew (@andrlfv) profile picture

Founder and CEO @OnyonSystems building the future of AI. Generative Art at @monomonio

Earth
Joined September 2009
Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)Photo by Andrew (@andrlfv)
Andrew retweeted
X/Twitter has increased the limit of photos and videos per post from 4 to 20.
1,388
8,679
24,283
40,823
30,237,614
Andrew retweeted
MiniMax-H3-Image in 2 steps, realtime pDMD LoRA on MiniMax's 33B image model realtime editing on ZeroGPU LoRA + one-file ComfyUI checkpoint: huggingface.co/akhaliq/minim… app: huggingface.co/spaces/akhali…
30
56
10
483
58,741
Earlier this week Tristan published his top YouTube channels: thume.ca/2026/11/01/youtube-… Which looked excellent so clearly I had to turn it into TristanTube: karpathy.ai/tristan-tube/ 47K super nerdy videos +TV mode for laid back watching. (One-shotted with Claude Code in 3 minutes)
Weekend hack: HackerTube. I was annoyed with YouTube's recommendations so I had Claude filter down 43M comments over 20 years of HN to 6,454 videos that keep being recommended. *Significantly* better viewing experience, ~2 hours with Opus 5.5: hackertube.org/ - Excellent recs without any logins (state is kept in local browser session) - Thumbnail is a representative frame from middle of the video instead of the super duper A/B test hyper-optimized user uploads - Underlying data from the comment analysis as one zip file, point your Claude at it and ask for your own version - HackerTube Longs ✊ A bit more on the process: I use Claude Code (desktop app for Mac), Opus 5.5 on Medium effort as my default. I only talk to Claude and have it do everything, the only "manual work" I helped with in tandem was in the browser (reminiscent of MenuGen): 1) I had to create some kind of API key from Google Cloud UI to query YouTube for video metadata, 2) I booted up a VPS box for serving and 3) I bought the domain name. Once the box was up I gave Claude the IP and told it update, secure, harden, deploy and optimize the code, and it did the whole thing. Many awesome projects that would have normally taken weeks/months even can now be few-shotted for breakfast. What else are we not thinking of.
740
796
243
9,458
1,098,711
People DO judge a book by its cover
42
242
39
3,248
185,975
Jim Cramer is a legend
593
3,879
769
26,574
2,009,259
Today, we release the weights of a new and improved Le Chonk, advancing the open frontier for agentic coding and cybersecurity. 💻 Built to build: Le Chonk ranks #1 on @cognition’s FrontierCode among open models, and sets a new benchmark for Pareto efficiency on @artificialanalysis. 🛡 Built to defend: State-of-the-art in cyberdefense, put through the paces by the world's leading cyber vendors, including @CrowdStrike, @cloudflare, @rubrik, @zscaler, and others. Built for sovereignty: Make it yours. Download the weights on @huggingface , customize it with Mistral Forge, and deploy it wherever you need it. Read more from our team and customers 👇:
458
822
541
6,426
898,791
Andrew retweeted
Average tech bro lifestyle.
200
3,752
254
28,454
854,236
Andrew retweeted
Introducing 𝙻𝙵𝙼𝟹: our new generation of Liquid Foundation Models. With a new tri-hybrid architecture and new multimodal encoders, the LFM3 family is the ideal model suite for specialized intelligence. > 6 model instances, across text, audio, vision, and omni modalities, generative and decision models. > 𝙻𝙵𝙼𝟹-𝟷.𝟽𝙱: text, 256k context window. > 𝙻𝙵𝙼𝟹-𝚅𝙻-𝟸𝙱: vision-language model with video-input support for spatial understanding > 𝙻𝙵𝙼𝟹-𝙰𝚞𝚍𝚒𝚘-𝟸𝙱: tool calls, voice chat, reasoning in 18 languages. > 𝙻𝙵𝙼𝟹-𝙾𝚖𝚗𝚒-𝟸𝙱: an omni model combines vision, audio, and language. > 𝚍𝟷-𝟸𝙱 and 𝚍𝟷-𝙾𝚖𝚗𝚒-𝟸𝙱: decision models with typed outputs and calibrated probabilities in a single pass. Omni adds audio input. > fine-tune on 1 GPU. > a new tri-hybrid architecture > architecture optimized for 14 processor classes across GPUs, CPUs, and NPUs. > 𝙻𝙵𝙼𝟹-𝚅𝙻-𝟸𝙱 is a beast. It performs competitively with Gemma 4 12B (6x larger), and Gemini 3.5 Flash Lite, and GPT-5.6 Luna, with 1.9B parameters. > 𝙻𝙵𝙼𝟹-𝟷.𝟽𝙱 is #1 on the Pareto frontier of device intelligence 𝚙𝚒𝚙𝚎𝚝𝚝𝚎.𝚕𝚒𝚚𝚞𝚒𝚍.𝚊𝚒 Every modality for specialized intelligence. Pick the one that fits your use case, customize it, and run it on-device. 1/ 🧵
540
1,258
738
9,047
2,401,436
ICYMI: Today was basically the worse day of the year for cyber security incidents
14
558
20
5,987
71,929
Soon there will be open source fruit flies on Huggingface
7
221
7
2,980
47,636
Andrew retweeted
If on-chain transactions were fireworks.
1,000
1,039
245
9,112
1,041,310
i love how as you make models smarter they become more vegan
94
49
24
1,082
109,647
Andrew retweeted
Keep your eye on the donut and not on the hole. In memory of David Lynch, we will continue to explore the otherworldly and the unknown. We will focus not on the loss, but on what we gained from the years we shared this planet with you. We will see you in our dreams.
645
16,777
1,337
89,156
5,178,408
being honest about the small things is how you earn trust to lie about the big things
86
12
5
138
17,590
A billion seconds ago was 1993 (31 years ago). A trillion seconds ago was 30,000 B.C.
30
54
13
314
142,963
Andrew retweeted
Just received word from King Kai. Legendary manga creator and the father of Dragon Ball, Akira Toriyama, has passed through the Other World Check-In Station.
801
22,913
2,308
137,897
6,036,402
Quem investe seu feriado em aprendizado sai na frente :)
Obrigado @solangegueiros @chainlinkpt @jeffbeltrao pelo feriado extremamente produtivo. #ConstellationHackathon 🚀
2
1
26
918
Resumo Semanal Tezos ✨ Veja os principais assuntos da semana. Matéria completa e links em brasiltezos.com.br Arte de capa: LISBON #2 DYSTOPIC #6 Artista: @monomonio ☑️ @objktcom revela Beta da versão 2.0: um foco renovado na arte 👇
1
3
6
153
Triste ver o @meteoro_br dar notícia assim, sem conhecimento básico. É irresponsável e desrespeitoso, nivelando todos envolvidos à golpistas. PF, procurem especialistas p/ desmistificar assuntos q não entendem, já q dizem fazer jornalismo sério! youtu.be/XHG42oeDniE
2
3
9
696
"If you want to be wealthy, spend your time earning, learning, or relaxing. Outsource or ignore everything else." @naval
14
388
9
2,581
128,063