BI

bitnet b1.58 2B 4T

BitnetGGUF by microsoftUpdated 2025-12-17
Download on Hugging Face
Parameters2.41B
Context4K
Layers30
Architecturebitnet-b1.58
Smallest quant1.2 GB
Quick answer

bitnet b1.58 2B 4T needs about 2.3 GB of memory at its main file (a 1.2 GB file) with an 8K context window, so it fits on a 4 GB GPU.

BitNet b1.58 2B4T – 2‑billion parameter 1‑bit LLM

BitNet b1.58 2B4T is a 2‑billion parameter transformer model trained from scratch with native 1‑bit weights. It supports a maximum context length of 4096 tokens and is released under the MIT license.

Good at

  • Efficient inference with low memory footprint
  • Native 1‑bit quantization for speed
  • Compatible with bitnet.cpp for CPU use

Limits

  • No vision or multimodal support
  • Requires 8‑bit activation support
  • Context limited to 4096 tokens

Best for

  • Deploying lightweight LLMs on CPUs
  • Running inference on devices with limited VRAM
  • Testing 1‑bit quantization research

Who it’s for. Developers and researchers who need a small, efficient LLM for CPU inference. The model is ideal for edge deployments and experimentation with native 1‑bit quantization.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—1.2 GB 2.0 GB
Fits Get file →

For a 1.19 GB GGUF file, the model uses about 2.3 GB of RAM at 8 k context. If you have 4 GB of RAM, this quant is suitable. For tighter budgets, consider a higher‑bit quant if available, but it is not listed here.

What GPU do you need for bitnet b1.58 2B 4T?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is the main file (1.2 GB file, about 2.3 GB in use).

Specifications

Architecture from the GGUF header

Architecturebitnet-b1.58
Parameters2.41B
Layers30
Trained context4,096 tokens
KV cache (fp16)75 KB per token

Source

Quantized bymicrosoft
FamilyBitnet
Licensemit
Downloads (HF)24,353
Last updated2025-12-17

Frequently asked questions about bitnet b1.58 2B 4T

How much VRAM does bitnet b1.58 2B 4T need?

About 2.3 GB at its main file with an 8K context window. That is the 1.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run bitnet b1.58 2B 4T on an 8 GB GPU?

Yes, with the smaller files: the 1.2 GB file needs about 2.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of bitnet b1.58 2B 4T GGUF?

It depends on the quant: file is 1.2 GB.

What context length does bitnet b1.58 2B 4T support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 0.6 GB at 8K and 0.3 GB at 4K.

Where can I download bitnet b1.58 2B 4T GGUF?

The files are on Hugging Face in the microsoft/bitnet-b1.58-2B-4T-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does bitnet b1.58 2B 4T use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run BitNet b1.58 2B4T on 8 GB VRAM?

The GGUF file is 1.19 GB and requires about 2.3 GB of RAM at 8 k context, so it can run on 8 GB VRAM with headroom for other processes.

Does the model support vision tasks?

No, the model is a text‑only transformer with no vision or multimodal capabilities.

What license does it use?

The model is released under the MIT license, allowing free use and modification.

Similar models

Browse all →

More from microsoft

All by microsoft →

More Bitnet models

All Bitnet →

Other tiny models

All tiny models →