Swallow GGUF models
96 GGUF models in the Swallow family, quantized by 7 uploaders. Sizes run from 6.74B to 70.6B parameters. The longest trained context is 128K tokens.
About the Swallow GGUF models
Swallow is represented on ggufmodels.com by 96 GGUF models, quantized by 7 uploaders on Hugging Face. The most downloaded are Swallow-7b-NVE-instruct-hf, Swallow 13b Instruct v0.1 and Swallow-7b-hf. Parameter counts run from 6.74B to 70.6B, and the longest trained context window is 128K tokens.
A Q4_K_M file of a Swallow model is between 4.1 to 42.5 GB, which is the quant most people start with. Most files come from mradermacher (72), RichardErkhov (12) and TheBloke (6). The quants you will meet most often are Q2_K (94), Q4_K_S (93), Q4_K_M (92) and Q3_K_M (92), where the number is how many models offer that quant.
Frequently asked questions
Which Swallow GGUF should I download?
The most downloaded Swallow GGUF here is Swallow-7b-NVE-instruct-hf. 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 Swallow model need?
Q4_K_M files of Swallow models on this site range from 4.1 to 42.5 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 Swallow GGUF files come from?
The most common quantizers here are mradermacher, RichardErkhov and TheBloke. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.