Question 1

How much VRAM does AI21-Jamba2-Mini need?

Accepted Answer

AI21-Jamba2-Mini (51.6B) requires approximately 115.26865510940551 GB of VRAM at full precision (FP16), 57.6 GB at 8-bit quantization (Q8), and 28.8 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 AI21-Jamba2-Mini?

Accepted Answer

We have 9 compatible GPUs indexed for AI21-Jamba2-Mini, including MI300X, H200 SXM, B200, B100, MI325X and 4 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 AI21-Jamba2-Mini?

Accepted Answer

The cheapest GPU to run AI21-Jamba2-Mini in the cloud is the MI300X at $1.99/hr. Other affordable options include H200 SXM ($2.35/hr) and B200 ($5.49/hr). Prices reflect the lowest on-demand rate across cloud providers.

Question 4

Can AI21-Jamba2-Mini run on consumer hardware?

Accepted Answer

AI21-Jamba2-Mini requires approximately 115.26865510940551 GB of VRAM at FP16, and even with 4-bit quantization (Q4) it needs about 28.8 GB. This exceeds the 24 GB available on the highest-end consumer GPU (RTX 4090), so AI21-Jamba2-Mini requires datacenter or cloud GPUs.

Question 5

What quantization options are available for AI21-Jamba2-Mini?

Accepted Answer

AI21-Jamba2-Mini supports the following quantization levels: FP16 (full precision): ~115.26865510940551 GB, Q8 (8-bit): ~57.6 GB, Q6 (6-bit): ~43.2 GB, Q4 (4-bit): ~28.8 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. As a Mixture of Experts (MoE) model with 16 experts (2 active), AI21-Jamba2-Mini has a larger total weight size but only activates a subset of parameters per token, which can affect effective memory usage during inference.

GPU	VRAM	Best Quant	From	Rent
MI300X	192 GB	FP16	$1.99/hr	Rent
H200 SXM	141 GB	FP16	$2.35/hr	Rent
B200	192 GB	FP16	$5.49/hr	Rent
B100	192 GB	FP16	—
MI325X	256 GB	FP16	—
MI350X	288 GB	FP16	—
Gaudi 3	128 GB	Q8	—
DGX Spark (GB10)	128 GB	Q8	—
M3 Ultra (192 GB)	144 GB	FP16	—

AI21-Jamba2-Mini

VRAM Requirements

Compatible GPUs

AI21-Jamba2-Mini FAQ

VRAM requirements for AI21-Jamba2-Mini at different quantization levels
Quantization	VRAM Required
FP16	115.3 GB
Q8	57.6 GB
Q6	43.2 GB
Q4	28.8 GB