BI

bitnet_b1_58-xl

BitnetGGUF by BoscoTheDogUpdated 2024-06-23
Download on Hugging Face
Parameters1.27B
Context2K
Layers24
Architecturebitnet
Smallest quant1.6 GB
Quick answer

bitnet_b1_58-xl needs about 3.7 GB of memory at Q8_0 (a 1.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 1.6 GB (Q8_0) to 5.1 GB (F32).

bitnet_b1_58-xl is a 1.27B-parameter Bitnet model in GGUF format, quantized by BoscoTheDog. Its trained context length is 2K tokens. The repository has 2 quant files, from 1.6 GB (Q8_0) to 5.1 GB (F32).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_01.6 GB 2.5 GB
Fits Get file →
F325.1 GB 6.0 GB
Fits Get file →

What GPU do you need for bitnet_b1_58-xl?

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 Q8_0 (1.6 GB file, about 3.7 GB in use).
  • 6 GB VRAM: the best quant that fits is Q8_0 (1.6 GB file, about 3.7 GB in use).
  • 8 GB VRAM: the best quant that fits is F32 (5.1 GB file, about 7.2 GB in use).

Which bitnet_b1_58-xl quant should you choose?

Q8_0 (1.6 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturebitnet
Parameters1.27B
Layers24
Trained context2,048 tokens
KV cache (fp16)192 KB per token

Source

Quantized byBoscoTheDog
FamilyBitnet
Licensemit
Downloads (HF)475
Last updated2024-06-23

Frequently asked questions about bitnet_b1_58-xl

How much VRAM does bitnet_b1_58-xl need?

About 3.7 GB at Q8_0 with an 8K context window. That is the 1.6 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-xl on an 8 GB GPU?

Yes, with F32: the 5.1 GB file needs about 7.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of bitnet_b1_58-xl GGUF?

It depends on the quant: Q8_0 is 1.6 GB, F32 is 5.1 GB.

What context length does bitnet_b1_58-xl support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 1.6 GB at 8K and 0.4 GB at 2K.

Where can I download bitnet_b1_58-xl GGUF?

The files are on Hugging Face in the BoscoTheDog/bitnet_b1_58-xl_q8_0_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-xl 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.

Similar models

Browse all →

More from BoscoTheDog

All by BoscoTheDog →

More Bitnet models

All Bitnet →

Other tiny models

All tiny models →