StarCoder GGUF models
211 GGUF models in the StarCoder family, quantized by 65 uploaders. Sizes run from 7M to 42.1B parameters. The longest trained context is 40K tokens.
About the StarCoder GGUF models
StarCoder is represented on ggufmodels.com by 211 GGUF models, quantized by 65 uploaders on Hugging Face. The most downloaded are Dolphincoder Starcoder2 15b, StarCoder and Starcoder. Parameter counts run from 7M to 42.1B, and the longest trained context window is 40K tokens.
A Q4_K_M file of a StarCoder model is between 0.1 to 25.9 GB, which is the quant most people start with. Most files come from mradermacher (54), RichardErkhov (35) and tensorblock (13). The quants you will meet most often are Q4_K_M (163), Q5_K_M (143), Q2_K (142) and Q3_K_M (141), where the number is how many models offer that quant.
Frequently asked questions
Which StarCoder GGUF should I download?
The most downloaded StarCoder GGUF here is Dolphincoder Starcoder2 15b. 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 StarCoder model need?
Q4_K_M files of StarCoder models on this site range from 0.1 to 25.9 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 StarCoder GGUF files come from?
The most common quantizers here are mradermacher, RichardErkhov and tensorblock. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.