|
2.7TB of Unified VRAM and officially out of rack space for now
|
|
0
|
1
|
November 2, 2026
|
|
Newest addition to the homelab! Made it extra homelab-y by building a wooden rack instead of buying more rails because they seem wildly expensive
|
|
0
|
1
|
November 2, 2026
|
|
The official documentation mentions that it can run on 64GB of RAM, but that involves offloading a larger portion of the model to the SSD, which significantly slows down performance
|
|
0
|
1
|
November 1, 2026
|
|
Today, I’ve had to go through the hassle of migrating that stack of Spark units to a TP3/4 cluster architecture—which has only made my back pain even worse
|
|
0
|
2
|
November 1, 2026
|
|
The same agent in your terminal
|
|
0
|
1
|
November 1, 2026
|
|
Hey GitLair Hackers, Devs, and Self-Hosted Hardware Addicts,
|
|
0
|
1
|
November 1, 2026
|
|
Open-source coding agent. Any model. Free on Mac, Windows and Linux
|
|
0
|
1
|
November 1, 2026
|
|
Stop this is what apple's mac mini should have looked like from day one
|
|
0
|
2
|
November 1, 2026
|
|
Haarness fixing a failing checkout test: it found the commit that dropped an await, wrote the fix, and now waits for your OK before touching the file
|
|
0
|
1
|
November 1, 2026
|
|
Liquid AI's new 1.7B edge model dropped this week and modders havew already removed its refusals and squeezed it into a 1.1GB GGUF
|
|
0
|
1
|
November 1, 2026
|
|
RTX 5090s being dismantled and desoldered by the thousands for AI servers | Tom's Hardware
|
|
0
|
1
|
November 1, 2026
|
|
Best home hardware for an AI rig
|
|
0
|
1
|
November 1, 2026
|
|
People Selling AI cheap
|
|
0
|
1
|
November 1, 2026
|
|
Connect Claude & ChatGPT
|
|
0
|
1
|
November 1, 2026
|
|
Gorgon Halo + DeepSeek V4.1-Flash-1017 + Maya
|
|
0
|
1
|
November 1, 2026
|
|
Our local LLM setup is decided—for now! Based on our benchmarks, we'll use: Mac: Qwen/oMLX Spark ×3: DS4.1/Owen1030
|
|
0
|
2
|
October 31, 2026
|
|
What people think when I say LLM is my hobby
|
|
0
|
1
|
October 31, 2026
|
|
Qwen3.8-27B 3-bit, Ignis w/ RTX 5060 Ti DRAM 16GB
|
|
0
|
1
|
October 31, 2026
|
|
Are people actually making money with AI automation businesses, and can a complete beginner do it too?
|
|
0
|
1
|
October 31, 2026
|
|
Ai automation still worth to try in 2026 is it crowded?
|
|
0
|
1
|
October 31, 2026
|
|
Nvidia RTX 20, 30 and 40-series GPUs can now unlock their full AI performance thanks to a new UEFI-level tool
|
|
0
|
1
|
October 31, 2026
|
|
Codex has actually processed 106 billion tokens—that’s 106B tokens in total!
|
|
0
|
1
|
October 31, 2026
|
|
The architectural design of DeepSeek’s newly open-sourced `dsh-superagent` highlights a core objective for next-generation agent systems: delivering "much value" while requiring "little human attention."
|
|
0
|
1
|
October 31, 2026
|
|
The ultimate "neurosurgical" tool for the AI world has arrived
|
|
0
|
1
|
October 31, 2026
|
|
SpawnSweep – Identify development servers, emulators, and build artifacts left behind by AI coding agents; preview them before cleaning up
|
|
0
|
1
|
October 31, 2026
|
|
AgentionAI is best known on Hugging Face for the Gyro Qwen3.8-Flash-Next quants
|
|
0
|
1
|
October 31, 2026
|
|
🔥 Run Qwen 3.8 (27B) Locally on Your Mac
|
|
1
|
2
|
October 31, 2026
|
|
Finished AI Server Best Models?
|
|
1
|
3
|
October 31, 2026
|
|
Seeking local AI hardware advice
|
|
0
|
1
|
October 31, 2026
|
|
Time to diversify
|
|
0
|
1
|
October 31, 2026
|