SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.17764 · 2024

The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 2 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
Entropy-xcy/bitnet158 reimplementation 1 of 1
microsoft/bitblas pwc_unofficial 1 of 1
FunctionStatusWhere it lives
absmax_quantize Ran Entropy-xcy/bitnet158/bitnet158/bitlinear158.py
pointer only (licence: AFL-3.0) · get_code("2346bd33c6ee4a6a")
get_rasterization_code Ran microsoft/bitblas/bitblas/base/utils.py
code served (permissive licence) · get_code("bb6778a281ed0015")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recent research, such as BitNet, is paving the way for a new era of 1-bit Large Language Models (LLMs). In this work, we introduce a 1-bit LLM variant, namely BitNet b1.58, in which every single parameter (or weight) of the LLM is ternary {-1, 0, 1}. It matches the full-precision (i.e., FP16 or BF16) Transformer LLM with the same model size and training tokens in terms of both perplexity and end-task performance, while being significantly more cost-effective in terms of latency, memory, throughput, and energy consumption. More profoundly, the 1.58-bit LLM defines a new scaling law and recipe for training new generations of LLMs that are both high-performance and cost-effective. Furthermore, it enables a new computation paradigm and opens the door for designing specific hardware optimized for 1-bit LLMs.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2402.17764")
get_code_for_paper("2402.17764")
have("2402.17764")

Connect an agent — have() is free.