Hy3 GGUF size and VRAM requirements

License: apache-2.0 ⬇ 877,722 ❤ 18
Parameters298.79B
Context262,144

vcruz305/Hy3-GGUF is a very large language model with 298.79 billion parameters, built on the hy_v3 architecture. It is released under the apache-2.0 license and has been downloaded 877,722 times.

To run vcruz305/Hy3-GGUF locally at a 4,096-token context, its quantized versions need between 86.89 GB (IQ1_M, lowest quality) and 297.27 GB (Q8_0, highest quality) of memory, weights plus KV cache and a system margin included.

Available GGUF quantizations for vcruz305/Hy3-GGUF include IQ1_M, IQ2_M, Q2_K, IQ3_XXS, Q3_K_M, IQ4_XS, Q4_K_M, Q5_K_M, Q8_0. The model supports a native context length of up to 262,144 tokens; a longer context grows the KV cache and the memory needed.

→ 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
IQ1_M 2.46 Very low 85.45 GB 0.63 GB 86.89 GB Insufficient
IQ2_M 2.68 Low 93.14 GB 0.63 GB 94.57 GB Insufficient
Q2_K 2.91 Low 101.28 GB 0.63 GB 102.71 GB Insufficient
IQ3_XXS 3.14 Low 109.28 GB 0.63 GB 110.71 GB Insufficient
Q3_K_M 3.82 Fair 132.94 GB 0.63 GB 134.37 GB Insufficient
IQ4_XS 4.26 Good 148.22 GB 0.63 GB 149.65 GB Insufficient
Q4_K_M 4.85 Good 168.57 GB 0.63 GB 170.0 GB Insufficient
Q5_K_M 5.68 Very good 197.61 GB 0.63 GB 199.04 GB Insufficient
Q8_0 8.51 Excellent 295.84 GB 0.63 GB 297.27 GB Insufficient

KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.

Frequently asked questions

What kind of model is vcruz305/Hy3-GGUF?

vcruz305/Hy3-GGUF is a language model with 298.79 billion parameters, based on the hy_v3 architecture. It is released under the apache-2.0 license and distributed as GGUF files for local inference.

Can I run vcruz305/Hy3-GGUF on an 8 GB GPU?

No. vcruz305/Hy3-GGUF does not fit on an 8 GB GPU, even with the smallest quantization and system RAM offloading.

Can I run vcruz305/Hy3-GGUF on a 16 GB GPU?

No. vcruz305/Hy3-GGUF does not fit on a 16 GB GPU, even with the smallest quantization and system RAM offloading.

Can I run vcruz305/Hy3-GGUF on a 24 GB GPU?

No. vcruz305/Hy3-GGUF does not fit on a 24 GB GPU, even with the smallest quantization and system RAM offloading.

What context length does vcruz305/Hy3-GGUF support?

vcruz305/Hy3-GGUF supports a native context length of up to 262,144 tokens. A longer context grows the KV cache, so it increases the memory needed to run the model.

What is the best quantization for vcruz305/Hy3-GGUF?

For vcruz305/Hy3-GGUF, a strong default is Q4_K_M, which needs about 170.0 GB and keeps most of the quality while roughly halving the memory versus 8-bit. With VRAM to spare, Q5_K_M or Q6_K add a little more quality; if you are tight on memory, a smaller quantization still runs. Pick the highest quantization that fits your VRAM.