Loading...

Tag trends are in beta. Feedback? Thoughts? Email me at [email protected]

Compression for Go web servers in 2026: brotli vs zstd, shared dictionaries, streaming

Sanglard analyzes the video compression techniques of Silpheed (Sega CD, 1993)

How do you measure entropy across binary chunks before compression?

Adaptive compression for inverted index

Perceptual Image Codec: What Matters in Practical Learned Image Compression

Pokemon Gen2 compression myth

HDMI 2.1 Display Stream Compression (DSC) Ready for Amdgpu Linux Driver

Universal tool for the most optimal compression of files based on file type and user priority

Applying LZ77-style sequence compression and LZW substitution to LLM context reduction

I built an experimental alternative to .nii.gz using Zstd, chunked encoding, and ROI-aware compression

Native WebP v1.3.0 Compression Improvements in Pure Go!

Modern LZ Compression Part 2: FSE and Arithmetic Coding

TurboQuant model weight compression support added to Llamacpp

KV Cache Compression 900000x Beyond TurboQuant and Per-Vector Shannon Limit

Comparing compression tools

Metal Lossy Compression Format

TurboQuant KV Compression and SSD Expert Streaming for M5 Pro and IOS

HT-NRC: A high-throughput and noise-resilient lossless image compression architecture for deep-space CMOS cameras

simd-bp128 integer compression library

player2text – local audio/video transcription API with auto-compression (FastAPI + faster-whisper)

Hardware Image Compression

Taking a Look at Compression Algorithms – Moncef Abboud

JPEG Compression

Show HN: TurboQuant for vector search – 2-4 bit compression

Micron, SanDisk Stocks Tumble After Google Unveils AI Memory Compression Breakthrough

TurboQuant: Redefining AI efficiency with extreme compression

Writing Parallel File Compression with Memory Control

Google unveils TurboQuant, a new AI memory compression algorithm — and yes, the internet is calling it 'Pied Piper' | TechCrunch

Apply video compression on KV cache to 10,000x less error at Q4 quant

Google's TurboQuant AI-compression algorithm can reduce LLM memory usage by 6x

More →