Ideas & society2024 · shared on Hacker News

The Era of 1-bit LLMs: ternary parameters for cost-effective computing

Explore The Era of 1-bit LLMs: ternary parameters for cost-effective computing, its original source and Hacker News discussions. Discover community perspectives and save it to your reading list.

Read original

The conversation around the idea.

Original Hacker News threads. Different perspectives, questions, and things worth thinking about.

Showing up to 100 related threads. A paper may appear in a comment rather than the original submission.

Links found in the discussion

Up to 50 recent source links, in HN item order.

Ynmstoker ·

GPUs Go Brrr

1,104 points263 comments
Yfgfm ·

The Era of 1-bit LLMs: ternary parameters for cost-effective computing

1,040 points447 comments
Ynoch ·

Run DeepSeek R1 Dynamic 1.58-bit

767 points332 comments
Ymathgenius ·

Galois Theory

467 points234 comments
Ytiahura ·

Nvidia CEO Jensen Huang announces new AI chips: ‘We need bigger GPUs’

393 points314 comments
Yhomarp ·

Towards 1-bit Machine Learning Models

351 points157 comments
Yrx_tx ·

AMD Unveils Ryzen 9000 CPUs for Desktop, Zen 5

334 points292 comments
Ylykahb ·

Scalable MatMul-Free Language Modeling

205 points30 comments
YPaulHoule ·

Why can't transformers learn multiplication?

161 points107 comments
Ygeox ·

Researchers run high-performing LLM on the energy needed to power a lightbulb

147 points70 comments
Ylibpcap ·

Microsoft researchers developed a hyper-efficient AI model that can run on CPUs

146 points64 comments
Ymatt_d ·

FPGA Architecture for Deep Learning: Survey and Future Directions

128 points52 comments
Ypella ·

AMD's Freshly-Baked MI350: An Interview with the Chief Architect

128 points75 comments
YFromTheArchives ·

The Trinary Dream Endures

48 points74 comments
Ymodinfo ·

1-Bit Bonsai Image 4B Image Generation for Local Devices

33 points7 comments
Yredm ·

Microsoft BitNet: 100B Param 1-Bit model for local CPUs

25 points3 comments
Ynekofneko ·

Kimi Released Kimi K2.5, Open-Source Visual SOTA-Agentic Model

7 points2 comments
Ycafed00d ·

Show HN: A Small BitNet GPT in MLX

4 points0 comments
YJoshTriplett ·

How to pack ternary numbers in 8-bit bytes

1 points0 comments

One idea leads to another.

Explore more papers