Speculative KV coding: losslessly compressing KV cache by up to ~4× (fergusfinn.com) | HN Screenshots

YHacker Newsscreenshots

Screenshot of fergusfinn.com: Speculative KV coding: losslessly compressing KV cache by up to ~4×

First screen, captured 4 months ago.

Looks similar

Screenshot of minimaxir.com: The mysterious Hy3 LLM is topping OpenRouter Model Rankings by a large margin

The mysterious Hy3 LLM is topping OpenRouter Model Rankings by a large margin (minimaxir.com)

131 points by freediver 5 months ago | 103 comments

Screenshot of ziglang.org: Zig's new bitCast semantics and LLVM back end improvements

Zig's new bitCast semantics and LLVM back end improvements (ziglang.org)

266 points by kouosi 4 months ago | 135 comments

Screenshot of tomjohnell.com: LLMs can be exhausting

LLMs can be exhausting (tomjohnell.com)

256 points by tjohnell 7 months ago | 171 comments

Screenshot of jngiam.bearblog.dev: LLM function calls don't scale; code orchestration is simpler, more effective

LLM function calls don't scale; code orchestration is simpler, more effective (jngiam.bearblog.dev)

251 points by jngiam1 1 year ago | 90 comments

Screenshot of jola.dev: Running local models on an M4 with 24GB memory

Running local models on an M4 with 24GB memory (jola.dev)

525 points by shintoist 5 months ago | 156 comments

A blog post describing a setup for running local LLMs such as Qwen via LM Studio on an Apple M4 Mac with 24GB of memory.

local-llm apple-silicon qwen lm-studio

Screenshot of fergusfinn.com: Bringing Up DeepSeek-V4-Flash on AMD MI300X

Bringing Up DeepSeek-V4-Flash on AMD MI300X (fergusfinn.com)

107 points by kkm 4 months ago | 17 comments