FullstackSensei

• Intel's Crescent Island PCB Leaks, Showing a Massive Xe3P GPU, 16-Pin Connector, 160GB LPDDR5X as Intel Sidesteps the HBM Shortage

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 89 comments

• Roo Code hit 3 million installs. We're shutting it down to go all-in on Roomote.

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 116 comments

• Popularity of DDR3 motherboards is growing rapidly - VideoCardz.com

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 89 comments

• ggml: backend-agnostic tensor parallelism by JohannesGaessler · Pull Request #19378 · ggml-org/llama.cpp

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 47 comments

• Build script for llama.cpp for ROCm (including Mi50) using the Rock artifacts

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 6 comments

• Copper Price Surge - PC Hardware Gets Even More Expensive

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 5 comments

• Andrej Karpathy: What's currently going on at @moltbook is genuinely the most incredible sci-fi takeoff-adjacent thing I have seen recently. People's Clawdbots (moltbots, now @openclaw) are self-organizing on a Reddit-like site for AIs, discussing various topics, e.g. even how to speak privately

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 45 comments

• ByteDance-Seed/Stable-DiffCoder-8B-Instruct · Hugging Face

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 10 comments

• Big tech companies, now "DRAM beggars," are staying in Pangyo and Pyeongtaek, demanding "give us some supplies."

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 95 comments

• For the first time in 5 years, Nvidia will not announce any new GPUs at CES — company quashes RTX 50 Super rumors as AI expected to take center stage

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 204 comments

• Industry Update: Supermicro Policy on Standalone Motherboards Sales Discontinued — Spectrum Sourcing

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 72 comments

• A Zettabyte Scale Answer to the DRAM Shortage

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 3 comments

• Unsloth's Devstral-2 123B GGUF is cooking!

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 1 comments

• Micron Announces Exit from Crucial Consumer Business

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 190 comments

• Jeff Geerling does what Jeff Geerling does best: Quad Strix Halo cluster using Framework Desktop

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 75 comments

• FlashVSR: Towards Real-Time Diffusion-Based Streaming Video Super-Resolution

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 0 comments

• Intel launches $299 Arc Pro B50 with 16GB of memory, 'Project Battlematrix' workstations with 24GB Arc Pro B60 GPUs

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 313 comments

• The True Story of ZLUDA: How CUDA Can Run on AMD & Intel GPUs

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 8 comments

• Intel to launch Arc Pro B60 graphics card with 24GB memory at Computex - VideoCardz.com

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 52 comments

• Intel Xeon Clearwater Forest with 288 Cores on Intel 18A at Hot Chips 2025

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 10 comments

• GitHub - karpathy/rendergit: Render any git repo into a single static HTML page for humans or LLMs

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 14 comments

• Help vote for improved Vulkan performance in ik_llama.cpp

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 12 comments

• Cognition, maker of the AI coding agent Devin, acquires Windsurf

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 13 comments

• Nvidia to wind down CUDA support for Maxwell and Pascal

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 41 comments

• THIS is the Most Important GPU of 2025

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 5 comments

• Unsloth Llama 4 Scout Q4_K_XL at 18 tk/s on triple P40 using llama.cpp!

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 8 comments

• Qwen3 - a unsloth Collection

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 31 comments

• SmolBoi: watercooled 3x RTX 3090 FE & EPYC 7642 in O11D (with build pics)

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 29 comments

• OpenAI in talks to buy Windsurf for about $3 billion, Bloomberg News reports

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 27 comments

• Any draft model that works (well?) with the March release of QwQ-32B?

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 15 comments

• Some details on Project Digits from PNY presentation

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 128 comments

• DeepSeek-R1 on Quad P40 + Dual Xeon E5-2699v5

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 41 comments

• New Karpathy's video: How I use LLMs

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 20 comments

• GitHub Copilot: The agent awakens

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 19 comments

• Clayton Christensen: Disruptive innovation

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 3 comments

• Anthropic: ‘Please don’t use AI’

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 154 comments

• OpenAI plans to open an office in Germany | TechCrunch

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 3 comments

• DeepSeek’s Lessons for Chinese AI

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 2 comments

• DeepSeek-R1 2.51 bits on Quad P40 + Dual Xeon E5-2699v4

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 0 comments

• Quad P40 + Dual Xeon E5-2699V4

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 0 comments

• Meta is reportedly scrambling multiple ‘war rooms’ of engineers to figure out how DeepSeek’s AI is beating everyone else at a fraction of the price

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 498 comments

• Quad P40 build and benchmarks with Qwen-2.5-Coder-32B and Llama 3.1-Nemotron-70B

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 17 comments

• Call for broken donor cards or coolers in the EU

  Posted by FullstackSensei@reddit | LocalLLaMA | View on Reddit | 2 comments