Quantize mistral-7b-instruct-v0.2 with llamafile-0.8 F16

mistral-7b-instruct-v0.2.F16.llamafile

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a183de12a41cfd4e42b998d42680286d75508c058d279ecbc770734e1b26328b
-size 14509085165
+oid sha256:9cf83f569cbd1bd3b2b0922f9a9dec0dab0f0970afe3437e9ec041ede7da3608
+size 14511837923