Y Hacker News new | ask | show | jobs

user: volodia created: 2008-05-28 karma: 545 submissions: Mercury 2 on PinchBench: Diffusion LLM benchmarked on real OpenClaw agent tasks 2 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments Mercury 2: Best-in-class speed-optimized intelligence at 1,200 tok/SEC 1 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments Finetuning 3-Bit LLMs on Consumer GPUs by Integrating with Modular Quantizers 2 points | 0 comments LLMTune: 4-Bit finetuning of 65B LLAMA models on a single consumer GPU 3 points | 0 comments LLMTune: 4-Bit Finetuning of LLMs on a Consumer GPI 2 points | 0 comments 0 points | 0 comments Don't have a $5k MacBook to run LLAMA65B? MiniLLM runs LLMs on GPUs in <500 LOC 3 points | 2 comments