Question 1

How much VRAM does bloomz-1b1 need?

Accepted Answer

bloomz-1b1 (1.1B) requires approximately 2.3811630249023437 GB of VRAM at full precision (FP16), 1.2 GB at 8-bit quantization (Q8), and 0.6 GB at 4-bit quantization (Q4). The actual VRAM usage may be slightly higher due to KV cache and runtime overhead.

Question 2

What GPUs can run bloomz-1b1?

Accepted Answer

We have 10 compatible GPUs indexed for bloomz-1b1, including L4, M60, RTX A4000, T4, A40 and 5 more. Compatibility is determined by whether the GPU's VRAM can fit the model's memory footprint, potentially using quantization to reduce memory usage.

Question 3

What is the cheapest GPU to run bloomz-1b1?

Accepted Answer

The cheapest GPU to run bloomz-1b1 in the cloud is the L4 at $0.20/hr. Other affordable options include M60 ($0.23/hr) and RTX A4000 ($0.25/hr). Prices reflect the lowest on-demand rate across cloud providers.

Question 4

Can bloomz-1b1 run on consumer hardware?

Accepted Answer

Yes, bloomz-1b1 can run on consumer GPUs like the RTX A4000 (16 GB VRAM) at full precision (FP16). Other compatible consumer GPUs include RTX 3090 and RTX A4500.

Question 5

What quantization options are available for bloomz-1b1?

Accepted Answer

bloomz-1b1 supports the following quantization levels: FP16 (full precision): ~2.3811630249023437 GB, Q8 (8-bit): ~1.2 GB, Q6 (6-bit): ~0.9 GB, Q4 (4-bit): ~0.6 GB. Lower quantization reduces VRAM usage at the cost of some quality loss. Q8 offers a good balance between quality and memory savings, while Q4 provides maximum compression.

GPU	VRAM	Best Quant	From	Rent
L4	24 GB	FP16	$0.20/hr
M60	8 GB	FP16	$0.23/hr
RTX A4000	16 GB	FP16	$0.25/hr	Rent
T4	16 GB	FP16	$0.35/hr
A40	48 GB	FP16	$0.44/hr	Rent
RTX 3090	24 GB	FP16	$0.46/hr	Rent
RTX A4500	20 GB	FP16	$0.50/hr	Rent
P4	8 GB	FP16	$0.60/hr
L40S	48 GB	FP16	$0.86/hr	Rent
RTX 5090	32 GB	FP16	$0.99/hr	Rent

bloomz-1b1

VRAM Requirements

Compatible GPUs

bloomz-1b1 FAQ

VRAM requirements for bloomz-1b1 at different quantization levels
Quantization	VRAM Required
FP16	2.4 GB
Q8	1.2 GB
Q6	0.9 GB
Q4	0.6 GB