Speeding up LLM inference 2x times (possibly)
Here's a project I've been working on for the last few months. It's a new (I think) algorithm, that allows to adjust smoothly - and in real time - how many calculations you'd like to do during inference of an LLM model. It seems that it's possible to do just 20-25% of weight multiplications instead of all of them, and still get good inference results. I implemented it to run on M1/M2/M3 GPU. The mmul approximation itself can be pushed to run 2x fast before the quality of output collapses. The inference speed is just a bit faster than Llama.cpp's, because the rest of implementation…
In plain words
This project presents an algorithm for accelerating large language model inference on Apple Silicon chips. It reduces the number of weight multiplications needed during inference to 20-25% of the total while maintaining output quality, allowing users to adjust speed and accuracy in real time. The implementation achieves approximately 2x speedup in matrix multiplication operations and runs comparably to Llama.cpp, with potential for further optimization through improved development.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Here's a project I've been working on for the last few months. It's a new (I think) algorithm, that allows to adjust smoothly - and in real time - how many calculations you'd like to do during inference of an LLM model. It seems that it's possible to do just 20-25% of weight multiplications instead of all of them, and still get good inference results. I implemented it to run on M1/M2/M3 GPU. The mmul approximation itself can be pushed to run 2x fast before the quality of output collapses. The inference speed is just a bit faster than Llama.cpp's, because the rest of implementation could be better, but with a better development I think it can be a new method to speed up inference - in addition to quantization. You could call it ad-hoc model distillation :) You can change the speed / accuracy of a model at will, in real time. Oh, and as a side effect, the data format allows to also choose how much of the model you want to load into the memory. You can decide to skip say 10-20-40% of the least important weights. It's implemented for Mistral, it was also tested slightly on Mixtral and Llama. It's for FP16 for now, but Q8 is in the works. The algorithm is described here, and the implementation is open source. https://kolinko.github.io/effort/ I know these are bold claims, but I hope they survive the scrutiny :)
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, April 2024
the whole month →
Supabase▲2,328The Postgres developer platform is now generally available
Dev tools · 2024 · supabase.com
Cal.com Platform▲1,602Build your pixel-perfect booking experience with Atoms
Dev tools · 2024 · cal.com
deco.cx 2.0▲1,080Build web apps 10x faster with Deno, JSX, TS & Tailwind
Dev tools · 2024 · decocms.com
