North-Mini-Code-1.0 GGUF size and VRAM requirements

License: apache-2.0 ⬇ 40,589 ❤ 2
Parameters30.48B
Context500,000

bartowski/North-Mini-Code-1.0-GGUF is a very large code-focused language model with 30.48 billion parameters, built on the cohere2moe architecture. It is released under the apache-2.0 license and has been downloaded 40,589 times.

To run bartowski/North-Mini-Code-1.0-GGUF locally at a 4,096-token context, its quantized versions need between 3.58 GB (GGUF, lowest quality) and 60.28 GB (BF16, highest quality) of memory, weights plus KV cache and a system margin included.

For most users the best balance is GGUF, needing about 3.58 GB. That means bartowski/North-Mini-Code-1.0-GGUF fits entirely in the VRAM of a 6 GB GPU or larger, running fully on the GPU.

→ Guide: How much VRAM do you need?

GGUF file size and memory by quantization

Compare real GGUF weight sizes, estimated KV cache and total memory for Q4, Q5, Q8 and every quantization published in this repository.

Quant.Bits QualityWeights KVTotal Speed~Verdict
GGUF 0.03 Very low 0.11 GB 2.66 GB 3.58 GB 3517.6 t/s Fits in VRAM
IQ2_XXS 2.23 Very low 7.93 GB 2.66 GB 11.39 GB 6.3 t/s Offload
IQ2_XS 2.47 Very low 8.77 GB 2.66 GB 12.24 GB 5.7 t/s Offload
IQ2_S 2.52 Very low 8.95 GB 2.66 GB 12.41 GB 5.6 t/s Offload
IQ2_M 2.77 Low 9.82 GB 2.66 GB 13.29 GB 5.1 t/s Offload
Q2_K 2.91 Low 10.33 GB 2.66 GB 13.79 GB 4.8 t/s Offload
Q2_K_L 2.94 Low 10.45 GB 2.66 GB 13.91 GB 4.8 t/s Offload
IQ3_XXS 3.41 Fair 12.1 GB 2.66 GB 15.57 GB 4.1 t/s Offload
Q3_K_S 3.56 Fair 12.63 GB 2.66 GB 16.1 GB 4.0 t/s Offload
IQ3_XS 3.73 Fair 13.23 GB 2.66 GB 16.7 GB 3.8 t/s Offload
Q3_K_M 3.73 Fair 13.23 GB 2.66 GB 16.7 GB 3.8 t/s Offload
Q3_K_L 3.87 Fair 13.74 GB 2.66 GB 17.21 GB 3.6 t/s Offload
IQ3_M 3.9 Fair 13.84 GB 2.66 GB 17.31 GB 3.6 t/s Offload
Q3_K_XL 3.91 Fair 13.87 GB 2.66 GB 17.33 GB 3.6 t/s Offload
IQ4_XS 4.35 Good 15.44 GB 2.66 GB 18.91 GB 3.2 t/s Offload
IQ4_NL 4.59 Good 16.3 GB 2.66 GB 19.76 GB 3.1 t/s Offload
Q4_0 4.6 Good 16.32 GB 2.66 GB 19.78 GB 3.1 t/s Offload
Q4_K_S 4.74 Good 16.81 GB 2.66 GB 20.27 GB 3.0 t/s Offload
Q4_K_M 4.92 Good 17.46 GB 2.66 GB 20.92 GB 2.9 t/s Offload
Q4_K_L 4.95 Good 17.58 GB 2.66 GB 21.04 GB 2.8 t/s Offload
Q4_1 5.06 Very good 17.97 GB 2.66 GB 21.44 GB 2.8 t/s Offload
Q5_K_S 5.55 Very good 19.7 GB 2.66 GB 23.16 GB 2.5 t/s Offload
Q5_K_M 5.73 Very good 20.34 GB 2.66 GB 23.81 GB 2.5 t/s Offload
Q5_K_L 5.77 Very good 20.47 GB 2.66 GB 23.93 GB 2.4 t/s Offload
Q6_K 6.93 Excellent 24.59 GB 2.66 GB 28.05 GB Insufficient
Q6_K_L 6.96 Excellent 24.71 GB 2.66 GB 28.17 GB Insufficient
Q8_0 8.51 Excellent 30.21 GB 2.66 GB 33.67 GB Insufficient
BF16 16.01 Excellent 56.81 GB 2.66 GB 60.28 GB Insufficient

KV cache estimated (architecture unavailable). Speed is a rough estimate bounded by memory bandwidth.

Frequently asked questions

How much VRAM do you need to run bartowski/North-Mini-Code-1.0-GGUF?

You need about 3.58 GB of VRAM to run bartowski/North-Mini-Code-1.0-GGUF entirely on the GPU using the GGUF quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.

Can I run bartowski/North-Mini-Code-1.0-GGUF on an 8 GB GPU?

Yes. With 8 GB of VRAM you can run bartowski/North-Mini-Code-1.0-GGUF fully on the GPU using GGUF (about 3.58 GB).

Can I run bartowski/North-Mini-Code-1.0-GGUF on a 16 GB GPU?

Yes. With 16 GB of VRAM you can run bartowski/North-Mini-Code-1.0-GGUF fully on the GPU using IQ3_XXS (about 15.57 GB).

Can I run bartowski/North-Mini-Code-1.0-GGUF on a 24 GB GPU?

Yes. With 24 GB of VRAM you can run bartowski/North-Mini-Code-1.0-GGUF fully on the GPU using Q5_K_L (about 23.93 GB).

What is the best quantization for bartowski/North-Mini-Code-1.0-GGUF?

If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.