pull down to refresh

768GB of cheap Intel Optane DIMM memory sticks used to run 1T-parameter LLMwww.tomshardware.com/tech-industry/artificial-intelligence/enthusiast-runs-1-trillion-parameter-llm-from-768gb-of-intel-optane-dimm-memory-sticks-local-kimi-k2-5-install-achieved-roughly-4-tokens-per-second
381 sats \ 4 comments \ @zuspotirko 25 May AI tech
related
LLM Memorygrantslatton.com/llm-memory
299 sats \ 2 comments \ @carter 2 Jul 2025 AI
VRAM sizes are disappointing in 2025
405 sats \ 5 comments \ @zuspotirko 10 Feb 2025 tech
PrismML — Introducing Ternary Bonsai: Top Intelligence at 1.58 Bitsprismml.com/news/ternary-bonsai
390 sats \ 0 comments \ @DEADBEEF 16 Apr AI
Google's new Gemma 4 12B model is designed to run on any laptop with 16GB of RAMarstechnica.com/google/2026/06/googles-new-gemma-4-open-ai-model-is-sized-for-your-laptop/
261 sats \ 0 comments \ @0xbitcoiner 3 Jun AI
What are your most surprising personal uses of LLMs?
1533 sats \ 39 comments \ @k00b 3 May AI AskSN
My first game built entirely with a local LLM: 🐍⏪ Rewind Snakerolznz.github.io/rewind-snake
5447 sats \ 2 comments \ @rolznz 25 May AI
LLM weights small enough to fit in a bitcoin tx
4307 sats \ 2 comments \ @rijndael 2 Jun AI bitcoin
Memory Safety Without Lifetime Parameterswww.open-std.org/jtc1/sc22/wg21/docs/papers/2024/p3444r0.html
206 sats \ 0 comments \ @hn 21 Oct 2024 tech
Ranking LLMs - Sharing a Useful Spreadsheet
334 sats \ 2 comments \ @joseph_at_nostr_fan 21 Apr 2025 AI
Episode 236: Tassadar [Model training run starting Monday]
477 sats \ 1 comment \ @AtlantisPleb 11 Jun openagents
Making The Smallest And Dumbest LLM With Extreme Quantizationhackaday.com/2025/10/23/making-the-smallest-and-dumbest-llm-with-extreme-quantization/
230 sats \ 0 comments \ @0xbitcoiner 24 Oct 2025 AI
Apple wants AI to run directly on its hardware instead of in the cloudarstechnica.com/apple/2023/12/apple-wants-ai-to-run-directly-on-its-hardware-instead-of-in-the-cloud/
226 sats \ 0 comments \ @DEADBEEF 22 Dec 2023 AI
What if... someone optimized a model for taking action
1915 sats \ 1 comment \ @optimism 3 Jul 2025 AI
Intel Discontinues Bitcoin-Mining Blockscale Chips, No Future Gens Announcedwww.tomshardware.com/news/intel-discontinues-bitcoin-mining-blockscale-chips-no-future-gens-announced
853 sats \ 5 comments \ @edblock 18 Apr 2023 bitcoin
Drivel-ology: Challenging LLMs with Interpreting Nonsense with Deptharxiv.org/abs/2509.03867
336 sats \ 0 comments \ @optimism 7 Sep 2025 AI
The inflation spark that could become a deflation shock?bondvigilantes.com/blog/2026/02/the-inflation-spark-that-could-become-a-deflation-shock/
264 sats \ 1 comment \ @0xbitcoiner 27 Feb tech econ
AMD Believes Unified Memory Architectures Open Up a "World of Possibilities"wccftech.com/amd-unified-memory-architectures-open-up-a-world-of-possibilities-shape-product-roadmaps/
527 sats \ 0 comments \ @zuspotirko 7 Jun AI tech
Understanding the LLM Bubbleamericanaffairsjournal.org/2026/02/understanding-the-llm-bubble/
384 sats \ 1 comment \ @co574 2 May AI
Pebble Index 01 - External Memory For Your Brainrepebble.com/index
281 sats \ 5 comments \ @Scoresby 10 Dec 2025 tech
The Sphere in Las Vegas has 4PB flash memory at 400 GB/s to deliver 16K contentwww.tomshardware.com/tech-industry/the-sphere-in-las-vegas-features-4pb-of-flash-memory-at-400-gbs-to-deliver-16k-content
768 sats \ 2 comments \ @DEADBEEF 17 Mar 2024 tech
US considering more than $10 billion in subsidies for Intel, Bloomberg reportswww.reuters.com/technology/us-considering-more-than-10-bln-subsidies-intel-bloomberg-news-2024-02-16/
467 sats \ 3 comments \ @Bitman 18 Feb 2024 tech freebie