Orion GGUF models
70 GGUF models in the Orion family, quantized by 19 uploaders. Sizes run from 1.79B to 25.2B parameters. The longest trained context is 256K tokens.
About the Orion GGUF models
Orion is represented on ggufmodels.com by 70 GGUF models, quantized by 19 uploaders on Hugging Face. The most downloaded are Orion 26B A4B v1.1, Orion 26B A4B v1 and Orion Kitchoon 26B A4B. Parameter counts run from 1.79B to 25.2B, and the longest trained context window is 256K tokens.
A Q4_K_M file of a Orion model is between 1.6 to 18.0 GB, which is the quant most people start with. Most files come from mradermacher (32), BeaverAI (8) and demonsu (5). The quants you will meet most often are Q6_K (52), Q4_K_M (50), Q2_K (48) and Q5_K_M (47), where the number is how many models offer that quant.
Frequently asked questions
Which Orion GGUF should I download?
The most downloaded Orion GGUF here is Orion 26B A4B v1.1. 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 Orion model need?
Q4_K_M files of Orion models on this site range from 1.6 to 18.0 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 Orion GGUF files come from?
The most common quantizers here are mradermacher, BeaverAI and demonsu. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.