QuIP#: 2-bit Quantization for LLMs (cornell-relaxml.github.io)

Screenshot of cornell-relaxml.github.io: QuIP#: 2-bit Quantization for LLMs

First screen, captured 2 years ago.

Looks similar

Screenshot of llasatts.github.io: Llasa: Llama-Based Speech Synthesis

Llasa: Llama-Based Speech Synthesis (llasatts.github.io)

156 points by CalmStorm 1 year ago | 18 comments

Screenshot of fuglede.github.io: Llama.ttf: A font which is also an LLM

Llama.ttf: A font which is also an LLM (fuglede.github.io)

554 points by fuglede_ 2 years ago | 142 comments

Project page for llama.ttf, a font file that embeds a Llama language model and inference engine via HarfBuzz's Wasm shaper.

llama harfbuzz webassembly fonts

Screenshot of wiki.c2.com: Tuple Space (2014)

Tuple Space (2014)(wiki.c2.com)

98 points by paulgb 3 years ago | 50 comments

Screenshot of jadidbourbaki.github.io: Faster prompt lookup drafting in llama.cpp

Faster prompt lookup drafting in llama.cpp (jadidbourbaki.github.io)

72 points by pptadversary 29 days ago | 11 comments

Blog post by Hayder Tirmazi describing optimizations that make prompt lookup drafting in llama.cpp up to 42x faster (140x with Daniel Lemire's follow-up PR) with less memory.

llama-cpp speculative-decoding prompt-lookup-decoding llm-inference

Screenshot of bellard.org: Ts_zip: Text Compression Using Large Language Models

Ts_zip: Text Compression Using Large Language Models (bellard.org)

224 points by Deeg9rie9usi 3 years ago | 126 comments

Screenshot of justine.lol: LLaMA now goes faster on CPUs

LLaMA now goes faster on CPUs (justine.lol)

1286 points by lawrencechen 2 years ago | 413 comments

Justine Tunney's blog post describing 84 new matrix multiplication kernels that make llamafile prompt evaluation 30-500% faster on CPUs than llama.cpp.

llamafile llama-cpp matrix-multiplication cpu-inference

Looks similar

Llasa: Llama-Based Speech Synthesis(llasatts.github.io)

Llama.ttf: A font which is also an LLM(fuglede.github.io)

Tuple Space (2014)(wiki.c2.com)

Faster prompt lookup drafting in llama.cpp(jadidbourbaki.github.io)

Ts_zip: Text Compression Using Large Language Models(bellard.org)

LLaMA now goes faster on CPUs(justine.lol)

Llasa: Llama-Based Speech Synthesis (llasatts.github.io)

Llama.ttf: A font which is also an LLM (fuglede.github.io)

Faster prompt lookup drafting in llama.cpp (jadidbourbaki.github.io)

Ts_zip: Text Compression Using Large Language Models (bellard.org)

LLaMA now goes faster on CPUs (justine.lol)