Neural GGUF models
104 GGUF models in the Neural family, quantized by 34 uploaders. Sizes run from 1.84B to 46.7B parameters. The longest trained context is 1024K tokens.
About the Neural GGUF models
Neural is represented on ggufmodels.com by 104 GGUF models, quantized by 34 uploaders on Hugging Face. The most downloaded are intel_neural-chat-7b-v3-2, intel_neural-chat-7b-v3-1 and Neural AlphaMistral 7B. Parameter counts run from 1.84B to 46.7B, and the longest trained context window is 1024K tokens.
A Q4_K_M file of a Neural model is between 1.2 to 28.6 GB, which is the quant most people start with. Most files come from mradermacher (42), tensorblock (9) and featherless-ai-quants (6). The quants you will meet most often are Q2_K (80), Q4_K_M (78), Q5_K_M (77) and Q6_K (76), where the number is how many models offer that quant.
Frequently asked questions
Which Neural GGUF should I download?
The most downloaded Neural GGUF here is intel_neural-chat-7b-v3-2. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.
How much VRAM does a Neural model need?
Q4_K_M files of Neural models on this site range from 1.2 to 28.6 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.
Where do Neural GGUF files come from?
The most common quantizers here are mradermacher, tensorblock and featherless-ai-quants. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.