BL

Bigscience.Bloom

BloomGGUF by DevQuasar-13Updated 2025-03-17
Download on Hugging Face
Parameters180B
Context14K
Layers70
Architecturebloom
Smallest quant68.1 GB
Quick answer

Bigscience.Bloom needs about 148.1 GB of memory at Q4_K_M (a 114.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 6 quants, from 68.1 GB (Q2_K) to 191.1 GB (Q8_0).

Bigscience.Bloom is a 180B-parameter Bloom model in GGUF format, quantized by DevQuasar-13. Its trained context length is 14K tokens. The repository has 6 quant files, from 68.1 GB (Q2_K) to 191.1 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K68.1 GB (5 parts) 101.5 GB
Too big Get file →
Q3_K_M94.4 GB (7 parts) 127.8 GB
Too big Get file →
Q4_K_M114.7 GB (9 parts) 148.1 GB
Too big Get file →
Q5_K_M133.6 GB (10 parts) 167.0 GB
Too big Get file →
Q6_K147.6 GB (11 parts) 181.0 GB
Too big Get file →
Q8_0191.1 GB (14 parts) 224.5 GB
Too big Get file →

What GPU do you need for Bigscience.Bloom?

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: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Which Bigscience.Bloom quant should you choose?

Q4_K_M is the usual starting point for Bigscience.Bloom: it balances file size and quality well. Here it is a 114.7 GB file.

The Q8_0 file is 191.1 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 68.1 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturebloom
Parameters180B
Layers70
Trained context14,336 tokens
KV cache (fp16)3920 KB per token

Source

Quantized byDevQuasar-13
FamilyBloom
Downloads (HF)305
Last updated2025-03-17

Frequently asked questions about Bigscience.Bloom

How much VRAM does Bigscience.Bloom need?

About 148.1 GB at Q4_K_M with an 8K context window. That is the 114.7 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 Bigscience.Bloom on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 101.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Bigscience.Bloom GGUF?

It depends on the quant: Q2_K is 68.1 GB, Q3_K_M is 94.4 GB, Q4_K_M is 114.7 GB, Q5_K_M is 133.6 GB, Q6_K is 147.6 GB, Q8_0 is 191.1 GB.

What context length does Bigscience.Bloom support?

The GGUF header lists a trained context length of 14,336 tokens (14K). The KV cache grows with the context you use: about 32.9 GB at 8K and 57.5 GB at 14K.

Where can I download Bigscience.Bloom GGUF?

The files are on Hugging Face in the DevQuasar-13/bigscience.bloom-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 Bigscience.Bloom use?

Hugging Face lists the license as bigscience-bloom-rail-1.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Other GGUFs of bloom

Similar models

Browse all →

More from DevQuasar-13

All by DevQuasar-13 →

More Bloom models

All Bloom →

Other large models

All large models →