Add Q4_K_M GGUF quantization (4.7GB)

Recommended for production use with llama.cpp and Ollama.
Maintains full quality while being 3.4x smaller than FP16.

.gitattributes

@@ -8,3 +8,4 @@
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gguf/*.gguf filter=lfs diff=lfs merge=lfs -text

gguf/wraith-8b-Q4_K_M.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6e77450d8073013067365d97cdbf5d24b036588288381c92caf165e444fb2bb
+size 4920738720