Question 1

How much VRAM does North-Mini-Code-1.0-w4a16 need?

Accepted Answer

North-Mini-Code-1.0-w4a16 (30.5B) requires approximately 70.64635162353515 GB of VRAM at full precision (FP16), 35.3 GB at 8-bit quantization (Q8), and 17.7 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 North-Mini-Code-1.0-w4a16?

Accepted Answer

We have 10 compatible GPUs indexed for North-Mini-Code-1.0-w4a16, including A100 80GB PCIe, A100 80GB, H100 MEGA, H100 SXM, H100 NVL 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 North-Mini-Code-1.0-w4a16?

Accepted Answer

The cheapest GPU to run North-Mini-Code-1.0-w4a16 in the cloud is the A100 80GB PCIe at $1.39/hr. Other affordable options include A100 80GB ($1.39/hr) and H100 MEGA ($1.80/hr). Prices reflect the lowest on-demand rate across cloud providers.

Question 4

Can North-Mini-Code-1.0-w4a16 run on consumer hardware?

Accepted Answer

Yes, North-Mini-Code-1.0-w4a16 can run on consumer GPUs like the RTX Pro 6000 Blackwell (96 GB VRAM) at full precision (FP16).

Question 5

What quantization options are available for North-Mini-Code-1.0-w4a16?

Accepted Answer

North-Mini-Code-1.0-w4a16 supports the following quantization levels: FP16 (full precision): ~70.64635162353515 GB, Q8 (8-bit): ~35.3 GB, Q6 (6-bit): ~26.5 GB, Q4 (4-bit): ~17.7 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 128 experts (8 active), North-Mini-Code-1.0-w4a16 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
A100 80GB PCIe	80 GB	Q8	$1.39/hr	Rent
A100 80GB	80 GB	Q8	$1.39/hr	Rent
H100 MEGA	80 GB	Q8	$1.80/hr	Rent
H100 SXM	80 GB	Q8	$1.80/hr	Rent
H100 NVL	94 GB	FP16	$1.80/hr	Rent
H100 PCIe	80 GB	Q8	$1.80/hr	Rent
RTX Pro 6000 Blackwell	96 GB	FP16	$1.99/hr	Rent
MI300X	192 GB	FP16	$2.19/hr	Rent
GH200	96 GB	FP16	$2.29/hr
H200 SXM	141 GB	FP16	$2.35/hr	Rent

North-Mini-Code-1.0-w4a16

VRAM Requirements

Compatible GPUs

North-Mini-Code-1.0-w4a16 FAQ

VRAM requirements for North-Mini-Code-1.0-w4a16 at different quantization levels
Quantization	VRAM Required
FP16	70.6 GB
Q8	35.3 GB
Q6	26.5 GB
Q4	17.7 GB