bartowski commited on
Commit
ce24317
0 Parent(s):

Duplicate from bartowski/Meta-Llama-3-70B-Instruct-GGUF

Browse files
Files changed (33) hide show
  1. .gitattributes +66 -0
  2. Meta-Llama-3-70B-Instruct-IQ1_M.gguf +3 -0
  3. Meta-Llama-3-70B-Instruct-IQ1_S.gguf +3 -0
  4. Meta-Llama-3-70B-Instruct-IQ2_M.gguf +3 -0
  5. Meta-Llama-3-70B-Instruct-IQ2_S.gguf +3 -0
  6. Meta-Llama-3-70B-Instruct-IQ2_XS.gguf +3 -0
  7. Meta-Llama-3-70B-Instruct-IQ2_XXS.gguf +3 -0
  8. Meta-Llama-3-70B-Instruct-IQ3_M.gguf +3 -0
  9. Meta-Llama-3-70B-Instruct-IQ3_S.gguf +3 -0
  10. Meta-Llama-3-70B-Instruct-IQ3_XS.gguf +3 -0
  11. Meta-Llama-3-70B-Instruct-IQ3_XXS.gguf +3 -0
  12. Meta-Llama-3-70B-Instruct-IQ4_NL.gguf +3 -0
  13. Meta-Llama-3-70B-Instruct-IQ4_XS.gguf +3 -0
  14. Meta-Llama-3-70B-Instruct-Q2_K.gguf +3 -0
  15. Meta-Llama-3-70B-Instruct-Q3_K_L.gguf +3 -0
  16. Meta-Llama-3-70B-Instruct-Q3_K_M.gguf +3 -0
  17. Meta-Llama-3-70B-Instruct-Q3_K_S.gguf +3 -0
  18. Meta-Llama-3-70B-Instruct-Q4_K_M.gguf +3 -0
  19. Meta-Llama-3-70B-Instruct-Q4_K_S.gguf +3 -0
  20. Meta-Llama-3-70B-Instruct-Q5_K_M.gguf +3 -0
  21. Meta-Llama-3-70B-Instruct-Q5_K_S.gguf +3 -0
  22. Meta-Llama-3-70B-Instruct-Q6_K.gguf/Meta-Llama-3-70B-Instruct-Q6_K-00001-of-00002.gguf +3 -0
  23. Meta-Llama-3-70B-Instruct-Q6_K.gguf/Meta-Llama-3-70B-Instruct-Q6_K-00002-of-00002.gguf +3 -0
  24. Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00001-of-00003.gguf +3 -0
  25. Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00002-of-00003.gguf +3 -0
  26. Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00003-of-00003.gguf +3 -0
  27. Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00001-of-00005.gguf +3 -0
  28. Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00002-of-00005.gguf +3 -0
  29. Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00003-of-00005.gguf +3 -0
  30. Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00004-of-00005.gguf +3 -0
  31. Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00005-of-00005.gguf +3 -0
  32. Meta-Llama-3-70B-Instruct.imatrix +3 -0
  33. README.md +303 -0
.gitattributes ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ Meta-Llama-3-70B-Instruct-IQ1_M.gguf filter=lfs diff=lfs merge=lfs -text
37
+ Meta-Llama-3-70B-Instruct-IQ1_S.gguf filter=lfs diff=lfs merge=lfs -text
38
+ Meta-Llama-3-70B-Instruct-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
39
+ Meta-Llama-3-70B-Instruct-IQ2_S.gguf filter=lfs diff=lfs merge=lfs -text
40
+ Meta-Llama-3-70B-Instruct-IQ2_XS.gguf filter=lfs diff=lfs merge=lfs -text
41
+ Meta-Llama-3-70B-Instruct-IQ2_XXS.gguf filter=lfs diff=lfs merge=lfs -text
42
+ Meta-Llama-3-70B-Instruct-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
43
+ Meta-Llama-3-70B-Instruct-IQ3_S.gguf filter=lfs diff=lfs merge=lfs -text
44
+ Meta-Llama-3-70B-Instruct-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
45
+ Meta-Llama-3-70B-Instruct-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text
46
+ Meta-Llama-3-70B-Instruct-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
47
+ Meta-Llama-3-70B-Instruct-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
48
+ Meta-Llama-3-70B-Instruct-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
49
+ Meta-Llama-3-70B-Instruct-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
50
+ Meta-Llama-3-70B-Instruct-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
51
+ Meta-Llama-3-70B-Instruct-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
52
+ Meta-Llama-3-70B-Instruct-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
53
+ Meta-Llama-3-70B-Instruct-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
54
+ Meta-Llama-3-70B-Instruct-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
55
+ Meta-Llama-3-70B-Instruct-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
56
+ Meta-Llama-3-70B-Instruct-Q6_K.gguf/Meta-Llama-3-70B-Instruct-Q6_K-00001-of-00002.gguf filter=lfs diff=lfs merge=lfs -text
57
+ Meta-Llama-3-70B-Instruct-Q6_K.gguf/Meta-Llama-3-70B-Instruct-Q6_K-00002-of-00002.gguf filter=lfs diff=lfs merge=lfs -text
58
+ Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00001-of-00003.gguf filter=lfs diff=lfs merge=lfs -text
59
+ Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00002-of-00003.gguf filter=lfs diff=lfs merge=lfs -text
60
+ Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00003-of-00003.gguf filter=lfs diff=lfs merge=lfs -text
61
+ Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00001-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
62
+ Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00002-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
63
+ Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00003-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
64
+ Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00004-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
65
+ Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00005-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
66
+ Meta-Llama-3-70B-Instruct.imatrix filter=lfs diff=lfs merge=lfs -text
Meta-Llama-3-70B-Instruct-IQ1_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c59effcfec3292e466a938d66801c4ae0b4ae7806f90dde3b190617fb9fe84d6
3
+ size 16751195936
Meta-Llama-3-70B-Instruct-IQ1_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63ac9ca8a01ed0d7d2df385b2ec5cfa5513bf2106876d26372d3bd7af63e3b49
3
+ size 15343482656
Meta-Llama-3-70B-Instruct-IQ2_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:190bbc50f4f5e95193400872618813063c5d538bbd7d7bd8df89adc12ac8d29d
3
+ size 24119293728
Meta-Llama-3-70B-Instruct-IQ2_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0812d186c762b04a16115b8b22254a7367c0ce3c311450d27c3123a2fed7d85
3
+ size 22242342688
Meta-Llama-3-70B-Instruct-IQ2_XS.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6738130235f2bda476b7e499f9d9036971a9b0a1ce0b9b9798946307f8f38cdd
3
+ size 21142107936
Meta-Llama-3-70B-Instruct-IQ2_XXS.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe41e8a16c0c756a178581e7b343bd9b1275fb6fb8ab13bd9e0ea3e35535ffd8
3
+ size 19097384736
Meta-Llama-3-70B-Instruct-IQ3_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffe9f3e1f42c7cea2a6abbba2e82ee743ccea88588e98b3dd697f625ec42f2e3
3
+ size 31937034016
Meta-Llama-3-70B-Instruct-IQ3_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5271e6839c8e43d721b8b3b76cb81bcd9ee0c17f4380cd4bb2531b00157ada61
3
+ size 30912050976
Meta-Llama-3-70B-Instruct-IQ3_XS.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a71c35b14669ef117e8e104369303959be0187c53f8268e41d3d2a0d7db0d9d4
3
+ size 29307729696
Meta-Llama-3-70B-Instruct-IQ3_XXS.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9889fbcf9b8b109010153c3681c5e39072ca6f83c426a6725182791ebc506659
3
+ size 27469494048
Meta-Llama-3-70B-Instruct-IQ4_NL.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc9e432dd3f2e987f041dceb4355a4c61aa31ebbda16e3fce06af7d66b3a01a6
3
+ size 40053618464
Meta-Llama-3-70B-Instruct-IQ4_XS.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f72ecf45d624a03f04dbf7390fb5b3b1084b2e8a31dd3c4f8d1bb888ff151b2
3
+ size 37902661408
Meta-Llama-3-70B-Instruct-Q2_K.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5235341296d2b33566a339ab36ba2f83c45cbb8ac3551ba575130b07316b73a8
3
+ size 26375108384
Meta-Llama-3-70B-Instruct-Q3_K_L.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6097059e2180c9ca39f521343ee82f90011ded8ebf1674554410ec221b2c1dd
3
+ size 37140592416
Meta-Llama-3-70B-Instruct-Q3_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5f3632c7dc4dec225939c0362742ebaab5110e22599f11b7a79606a25ea52bf
3
+ size 34267494176
Meta-Llama-3-70B-Instruct-Q3_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d324ea99786adefa63fef55c929c92f0567958cd1e78bffe2fe6ca19d8a00ef
3
+ size 30912050976
Meta-Llama-3-70B-Instruct-Q4_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30d3c22063a2f97191a5ba714b9a40b36866eb3b340ed964cec1d302008b1e39
3
+ size 42520393504
Meta-Llama-3-70B-Instruct-Q4_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e7d6082e0b97745072ec1fbeeceba3951f248555c9f2eb127b83cc2a612ec19
3
+ size 40347219744
Meta-Llama-3-70B-Instruct-Q5_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c90220d69cec3fd849d9f915af794583e688db16484cab839237a9306f21a06
3
+ size 49949816608
Meta-Llama-3-70B-Instruct-Q5_K_S.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95474bba2401407b4dcf7535066e82ca1272d2aedeb661973767df6770787c1f
3
+ size 48657446688
Meta-Llama-3-70B-Instruct-Q6_K.gguf/Meta-Llama-3-70B-Instruct-Q6_K-00001-of-00002.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:319a64df51fd3c9196ac02c081ee928fb877a0b1c6e84b2ab974f80f1cb0cafc
3
+ size 32141175808
Meta-Llama-3-70B-Instruct-Q6_K.gguf/Meta-Llama-3-70B-Instruct-Q6_K-00002-of-00002.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b31f40134a5bb3611f9eb53c6c9e15e942a472c65199025aa0b2dda1f55068b6
3
+ size 25746967520
Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00001-of-00003.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5deabbbc9db2290535ca4e014a4f782805941937867091995fa6acad7ec1d20b
3
+ size 32036361856
Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00002-of-00003.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0391d1331f1af45d21bbe2d067a8be0cc3aeeda15661ce0f8962d712fbadf009
3
+ size 32070912576
Meta-Llama-3-70B-Instruct-Q8_0.gguf/Meta-Llama-3-70B-Instruct-Q8_0-00003-of-00003.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:244385cedf0d51b23010da715fda1f117bea8a42173cfbfbb55c2d44234cacb6
3
+ size 10867775360
Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00001-of-00005.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c6c3121e4b5e13eb94837980293816ba1481a97a13e04449af4be939c6db398
3
+ size 32141537792
Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00002-of-00005.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d12d8a168192a4a07e5cb6d37ea0f31e575677dc0393b53e1bdcb13a5879869d
3
+ size 32045738016
Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00003-of-00005.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7973a04217ef4f8d721885b5c3b513f75d67c424121367ac8724dc5adb1aab5c
3
+ size 32045738016
Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00004-of-00005.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ca4f6c7d274ec24969538efadb19cd15ec3e0ea4960f4b1beba781893004517
3
+ size 31743682240
Meta-Llama-3-70B-Instruct-fp16.gguf/Meta-Llama-3-70B-Instruct-fp16-00005-of-00005.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3de1f454f0f75a769098bb20342cc390232546d6ccc220e195cc27a125f8481a
3
+ size 13141216992
Meta-Llama-3-70B-Instruct.imatrix ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc30db5582c459705521a6c493a0394ba48ed9fced9182319e1904dbd8a28b9d
3
+ size 24922295
README.md ADDED
@@ -0,0 +1,303 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ pipeline_tag: text-generation
5
+ tags:
6
+ - facebook
7
+ - meta
8
+ - pytorch
9
+ - llama
10
+ - llama-3
11
+ license: other
12
+ license_name: llama3
13
+ license_link: LICENSE
14
+ extra_gated_prompt: >-
15
+ ### META LLAMA 3 COMMUNITY LICENSE AGREEMENT
16
+
17
+ Meta Llama 3 Version Release Date: April 18, 2024
18
+
19
+ "Agreement" means the terms and conditions for use, reproduction, distribution and modification of the
20
+ Llama Materials set forth herein.
21
+
22
+ "Documentation" means the specifications, manuals and documentation accompanying Meta Llama 3
23
+ distributed by Meta at https://llama.meta.com/get-started/.
24
+
25
+ "Licensee" or "you" means you, or your employer or any other person or entity (if you are entering into
26
+ this Agreement on such person or entity’s behalf), of the age required under applicable laws, rules or
27
+ regulations to provide legal consent and that has legal authority to bind your employer or such other
28
+ person or entity if you are entering in this Agreement on their behalf.
29
+
30
+ "Meta Llama 3" means the foundational large language models and software and algorithms, including
31
+ machine-learning model code, trained model weights, inference-enabling code, training-enabling code,
32
+ fine-tuning enabling code and other elements of the foregoing distributed by Meta at
33
+ https://llama.meta.com/llama-downloads.
34
+
35
+ "Llama Materials" means, collectively, Meta’s proprietary Meta Llama 3 and Documentation (and any
36
+ portion thereof) made available under this Agreement.
37
+
38
+ "Meta" or "we" means Meta Platforms Ireland Limited (if you are located in or, if you are an entity, your
39
+ principal place of business is in the EEA or Switzerland) and Meta Platforms, Inc. (if you are located
40
+ outside of the EEA or Switzerland).
41
+
42
+ 1. License Rights and Redistribution.
43
+
44
+ a. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable and royalty-free
45
+ limited license under Meta’s intellectual property or other rights owned by Meta embodied in the Llama
46
+ Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the
47
+ Llama Materials.
48
+
49
+ b. Redistribution and Use.
50
+
51
+ i. If you distribute or make available the Llama Materials (or any derivative works
52
+ thereof), or a product or service that uses any of them, including another AI model, you shall (A) provide
53
+ a copy of this Agreement with any such Llama Materials; and (B) prominently display “Built with Meta
54
+ Llama 3” on a related website, user interface, blogpost, about page, or product documentation. If you
55
+ use the Llama Materials to create, train, fine tune, or otherwise improve an AI model, which is
56
+ distributed or made available, you shall also include “Llama 3” at the beginning of any such AI model
57
+ name.
58
+
59
+ ii. If you receive Llama Materials, or any derivative works thereof, from a Licensee as part
60
+ of an integrated end user product, then Section 2 of this Agreement will not apply to you.
61
+
62
+ iii. You must retain in all copies of the Llama Materials that you distribute the following
63
+ attribution notice within a “Notice” text file distributed as a part of such copies: “Meta Llama 3 is
64
+ licensed under the Meta Llama 3 Community License, Copyright © Meta Platforms, Inc. All Rights
65
+ Reserved.”
66
+
67
+ iv. Your use of the Llama Materials must comply with applicable laws and regulations
68
+ (including trade compliance laws and regulations) and adhere to the Acceptable Use Policy for the Llama
69
+ Materials (available at https://llama.meta.com/llama3/use-policy), which is hereby incorporated by
70
+ reference into this Agreement.
71
+
72
+ v. You will not use the Llama Materials or any output or results of the Llama Materials to
73
+ improve any other large language model (excluding Meta Llama 3 or derivative works thereof).
74
+
75
+ 2. Additional Commercial Terms. If, on the Meta Llama 3 version release date, the monthly active users
76
+ of the products or services made available by or for Licensee, or Licensee’s affiliates, is greater than 700
77
+ million monthly active users in the preceding calendar month, you must request a license from Meta,
78
+ which Meta may grant to you in its sole discretion, and you are not authorized to exercise any of the
79
+ rights under this Agreement unless or until Meta otherwise expressly grants you such rights.
80
+
81
+ 3. Disclaimer of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY
82
+ OUTPUT AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS” BASIS, WITHOUT WARRANTIES OF
83
+ ANY KIND, AND META DISCLAIMS ALL WARRANTIES OF ANY KIND, BOTH EXPRESS AND IMPLIED,
84
+ INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES OF TITLE, NON-INFRINGEMENT,
85
+ MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR
86
+ DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE LLAMA MATERIALS AND
87
+ ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND
88
+ RESULTS.
89
+
90
+ 4. Limitation of Liability. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF
91
+ LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING
92
+ OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL,
93
+ INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE BEEN ADVISED
94
+ OF THE POSSIBILITY OF ANY OF THE FOREGOING.
95
+
96
+ 5. Intellectual Property.
97
+
98
+ a. No trademark licenses are granted under this Agreement, and in connection with the Llama
99
+ Materials, neither Meta nor Licensee may use any name or mark owned by or associated with the other
100
+ or any of its affiliates, except as required for reasonable and customary use in describing and
101
+ redistributing the Llama Materials or as set forth in this Section 5(a). Meta hereby grants you a license to
102
+ use “Llama 3” (the “Mark”) solely as required to comply with the last sentence of Section 1.b.i. You will
103
+ comply with Meta’s brand guidelines (currently accessible at
104
+ https://about.meta.com/brand/resources/meta/company-brand/ ). All goodwill arising out of your use
105
+ of the Mark will inure to the benefit of Meta.
106
+
107
+ b. Subject to Meta’s ownership of Llama Materials and derivatives made by or for Meta, with
108
+ respect to any derivative works and modifications of the Llama Materials that are made by you, as
109
+ between you and Meta, you are and will be the owner of such derivative works and modifications.
110
+
111
+ c. If you institute litigation or other proceedings against Meta or any entity (including a
112
+ cross-claim or counterclaim in a lawsuit) alleging that the Llama Materials or Meta Llama 3 outputs or
113
+ results, or any portion of any of the foregoing, constitutes infringement of intellectual property or other
114
+ rights owned or licensable by you, then any licenses granted to you under this Agreement shall
115
+ terminate as of the date such litigation or claim is filed or instituted. You will indemnify and hold
116
+ harmless Meta from and against any claim by any third party arising out of or related to your use or
117
+ distribution of the Llama Materials.
118
+
119
+ 6. Term and Termination. The term of this Agreement will commence upon your acceptance of this
120
+ Agreement or access to the Llama Materials and will continue in full force and effect until terminated in
121
+ accordance with the terms and conditions herein. Meta may terminate this Agreement if you are in
122
+ breach of any term or condition of this Agreement. Upon termination of this Agreement, you shall delete
123
+ and cease use of the Llama Materials. Sections 3, 4 and 7 shall survive the termination of this
124
+ Agreement.
125
+
126
+ 7. Governing Law and Jurisdiction. This Agreement will be governed and construed under the laws of
127
+ the State of California without regard to choice of law principles, and the UN Convention on Contracts
128
+ for the International Sale of Goods does not apply to this Agreement. The courts of California shall have
129
+ exclusive jurisdiction of any dispute arising out of this Agreement.
130
+
131
+ ### Meta Llama 3 Acceptable Use Policy
132
+
133
+ Meta is committed to promoting safe and fair use of its tools and features, including Meta Llama 3. If you
134
+ access or use Meta Llama 3, you agree to this Acceptable Use Policy (“Policy”). The most recent copy of
135
+ this policy can be found at [https://llama.meta.com/llama3/use-policy](https://llama.meta.com/llama3/use-policy)
136
+
137
+ #### Prohibited Uses
138
+
139
+ We want everyone to use Meta Llama 3 safely and responsibly. You agree you will not use, or allow
140
+ others to use, Meta Llama 3 to:
141
+ 1. Violate the law or others’ rights, including to:
142
+ 1. Engage in, promote, generate, contribute to, encourage, plan, incite, or further illegal or unlawful activity or content, such as:
143
+ 1. Violence or terrorism
144
+ 2. Exploitation or harm to children, including the solicitation, creation, acquisition, or dissemination of child exploitative content or failure to report Child Sexual Abuse Material
145
+ 3. Human trafficking, exploitation, and sexual violence
146
+ 4. The illegal distribution of information or materials to minors, including obscene materials, or failure to employ legally required age-gating in connection with such information or materials.
147
+ 5. Sexual solicitation
148
+ 6. Any other criminal activity
149
+ 2. Engage in, promote, incite, or facilitate the harassment, abuse, threatening, or bullying of individuals or groups of individuals
150
+ 3. Engage in, promote, incite, or facilitate discrimination or other unlawful or harmful conduct in the provision of employment, employment benefits, credit, housing, other economic benefits, or other essential goods and services
151
+ 4. Engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or related professional practices
152
+ 5. Collect, process, disclose, generate, or infer health, demographic, or other sensitive personal or private information about individuals without rights and consents required by applicable laws
153
+ 6. Engage in or facilitate any action or generate any content that infringes, misappropriates, or otherwise violates any third-party rights, including the outputs or results of any products or services using the Llama Materials
154
+ 7. Create, generate, or facilitate the creation of malicious code, malware, computer viruses or do anything else that could disable, overburden, interfere with or impair the proper working, integrity, operation or appearance of a website or computer system
155
+ 2. Engage in, promote, incite, facilitate, or assist in the planning or development of activities that present a risk of death or bodily harm to individuals, including use of Meta Llama 3 related to the following:
156
+ 1. Military, warfare, nuclear industries or applications, espionage, use for materials or activities that are subject to the International Traffic Arms Regulations (ITAR) maintained by the United States Department of State
157
+ 2. Guns and illegal weapons (including weapon development)
158
+ 3. Illegal drugs and regulated/controlled substances
159
+ 4. Operation of critical infrastructure, transportation technologies, or heavy machinery
160
+ 5. Self-harm or harm to others, including suicide, cutting, and eating disorders
161
+ 6. Any content intended to incite or promote violence, abuse, or any infliction of bodily harm to an individual
162
+ 3. Intentionally deceive or mislead others, including use of Meta Llama 3 related to the following:
163
+ 1. Generating, promoting, or furthering fraud or the creation or promotion of disinformation
164
+ 2. Generating, promoting, or furthering defamatory content, including the creation of defamatory statements, images, or other content
165
+ 3. Generating, promoting, or further distributing spam
166
+ 4. Impersonating another individual without consent, authorization, or legal right
167
+ 5. Representing that the use of Meta Llama 3 or outputs are human-generated
168
+ 6. Generating or facilitating false online engagement, including fake reviews and other means of fake online engagement
169
+ 4. Fail to appropriately disclose to end users any known dangers of your AI system
170
+
171
+ Please report any violation of this Policy, software “bug,” or other problems that could lead to a violation
172
+ of this Policy through one of the following means:
173
+ * Reporting issues with the model: [https://github.com/meta-llama/llama3](https://github.com/meta-llama/llama3)
174
+ * Reporting risky content generated by the model:
175
+ developers.facebook.com/llama_output_feedback
176
+ * Reporting bugs and security concerns: facebook.com/whitehat/info
177
+ * Reporting violations of the Acceptable Use Policy or unlicensed uses of Meta Llama 3: [email protected]
178
+ extra_gated_fields:
179
+ First Name: text
180
+ Last Name: text
181
+ Date of birth: date_picker
182
+ Country: country
183
+ Affiliation: text
184
+ geo: ip_location
185
+ By clicking Submit below I accept the terms of the license and acknowledge that the information I provide will be collected stored processed and shared in accordance with the Meta Privacy Policy: checkbox
186
+ extra_gated_description: The information you provide will be collected, stored, processed and shared in accordance with the [Meta Privacy Policy](https://www.facebook.com/privacy/policy/).
187
+ extra_gated_button_content: Submit
188
+ widget:
189
+ - example_title: Winter holidays
190
+ messages:
191
+ - role: system
192
+ content: You are a helpful and honest assistant. Please, respond concisely and truthfully.
193
+ - role: user
194
+ content: Can you recommend a good destination for Winter holidays?
195
+ - example_title: Programming assistant
196
+ messages:
197
+ - role: system
198
+ content: You are a helpful and honest code and programming assistant. Please, respond concisely and truthfully.
199
+ - role: user
200
+ content: Write a function that computes the nth fibonacci number.
201
+ inference:
202
+ parameters:
203
+ max_new_tokens: 300
204
+ stop:
205
+ - <|end_of_text|>
206
+ - <|eot_id|>
207
+ quantized_by: bartowski
208
+ ---
209
+
210
+ ## Llamacpp imatrix Quantizations of Meta-Llama-3-70B-Instruct
211
+
212
+ Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b2777">b2777</a> for quantization.
213
+
214
+ Original model: https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct
215
+
216
+ All quants made using imatrix option with dataset provided by Kalomaze [here](https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384)
217
+
218
+ ## Prompt format
219
+
220
+ ```
221
+ <|begin_of_text|><|start_header_id|>system<|end_header_id|>
222
+
223
+ {system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>
224
+
225
+ {prompt}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
226
+
227
+
228
+ ```
229
+
230
+ ## Download a file (not the whole branch) from below:
231
+
232
+ | Filename | Quant type | File Size | Description |
233
+ | -------- | ---------- | --------- | ----------- |
234
+ | [Meta-Llama-3-70B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/tree/main/Meta-Llama-3-70B-Instruct-Q8_0.gguf) | Q8_0 | 74.97GB | Extremely high quality, generally unneeded but max available quant. |
235
+ | [Meta-Llama-3-70B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/tree/main/Meta-Llama-3-70B-Instruct-Q6_K.gguf) | Q6_K | 57.88GB | Very high quality, near perfect, *recommended*. |
236
+ | [Meta-Llama-3-70B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-Q5_K_M.gguf) | Q5_K_M | 49.94GB | High quality, *recommended*. |
237
+ | [Meta-Llama-3-70B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-Q5_K_S.gguf) | Q5_K_S | 48.65GB | High quality, *recommended*. |
238
+ | [Meta-Llama-3-70B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-Q4_K_M.gguf) | Q4_K_M | 42.52GB | Good quality, uses about 4.83 bits per weight, *recommended*. |
239
+ | [Meta-Llama-3-70B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-Q4_K_S.gguf) | Q4_K_S | 40.34GB | Slightly lower quality with more space savings, *recommended*. |
240
+ | [Meta-Llama-3-70B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ4_NL.gguf) | IQ4_NL | 40.05GB | Decent quality, slightly smaller than Q4_K_S with similar performance *recommended*. |
241
+ | [Meta-Llama-3-70B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ4_XS.gguf) | IQ4_XS | 37.90GB | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
242
+ | [Meta-Llama-3-70B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-Q3_K_L.gguf) | Q3_K_L | 37.14GB | Lower quality but usable, good for low RAM availability. |
243
+ | [Meta-Llama-3-70B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-Q3_K_M.gguf) | Q3_K_M | 34.26GB | Even lower quality. |
244
+ | [Meta-Llama-3-70B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ3_M.gguf) | IQ3_M | 31.93GB | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
245
+ | [Meta-Llama-3-70B-Instruct-IQ3_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ3_S.gguf) | IQ3_S | 30.91GB | Lower quality, new method with decent performance, recommended over Q3_K_S quant, same size with better performance. |
246
+ | [Meta-Llama-3-70B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-Q3_K_S.gguf) | Q3_K_S | 30.91GB | Low quality, not recommended. |
247
+ | [Meta-Llama-3-70B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ3_XS.gguf) | IQ3_XS | 29.30GB | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
248
+ | [Meta-Llama-3-70B-Instruct-IQ3_XXS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ3_XXS.gguf) | IQ3_XXS | 27.46GB | Lower quality, new method with decent performance, comparable to Q3 quants. |
249
+ | [Meta-Llama-3-70B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-Q2_K.gguf) | Q2_K | 26.37GB | Very low quality but surprisingly usable. |
250
+ | [Meta-Llama-3-70B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ2_M.gguf) | IQ2_M | 24.11GB | Very low quality, uses SOTA techniques to also be surprisingly usable. |
251
+ | [Meta-Llama-3-70B-Instruct-IQ2_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ2_S.gguf) | IQ2_S | 22.24GB | Very low quality, uses SOTA techniques to be usable. |
252
+ | [Meta-Llama-3-70B-Instruct-IQ2_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ2_XS.gguf) | IQ2_XS | 21.14GB | Very low quality, uses SOTA techniques to be usable. |
253
+ | [Meta-Llama-3-70B-Instruct-IQ2_XXS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ2_XXS.gguf) | IQ2_XXS | 19.09GB | Lower quality, uses SOTA techniques to be usable. |
254
+ | [Meta-Llama-3-70B-Instruct-IQ1_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ1_M.gguf) | IQ1_M | 16.75GB | Extremely low quality, *not* recommended. |
255
+ | [Meta-Llama-3-70B-Instruct-IQ1_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-70B-Instruct-GGUF/blob/main/Meta-Llama-3-70B-Instruct-IQ1_S.gguf) | IQ1_S | 15.34GB | Extremely low quality, *not* recommended. |
256
+
257
+ ## Downloading using huggingface-cli
258
+
259
+ First, make sure you have hugginface-cli installed:
260
+
261
+ ```
262
+ pip install -U "huggingface_hub[cli]"
263
+ ```
264
+
265
+ Then, you can target the specific file you want:
266
+
267
+ ```
268
+ huggingface-cli download bartowski/Meta-Llama-3-70B-Instruct-GGUF --include "Meta-Llama-3-70B-Instruct-Q4_K_M.gguf" --local-dir ./ --local-dir-use-symlinks False
269
+ ```
270
+
271
+ If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:
272
+
273
+ ```
274
+ huggingface-cli download bartowski/Meta-Llama-3-70B-Instruct-GGUF --include "Meta-Llama-3-70B-Instruct-Q8_0.gguf/*" --local-dir Meta-Llama-3-70B-Instruct-Q8_0 --local-dir-use-symlinks False
275
+ ```
276
+
277
+ You can either specify a new local-dir (Meta-Llama-3-70B-Instruct-Q8_0) or download them all in place (./)
278
+
279
+ ## Which file should I choose?
280
+
281
+ A great write up with charts showing various performances is provided by Artefact2 [here](https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9)
282
+
283
+ The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.
284
+
285
+ If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.
286
+
287
+ If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.
288
+
289
+ Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.
290
+
291
+ If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.
292
+
293
+ If you want to get more into the weeds, you can check out this extremely useful feature chart:
294
+
295
+ [llama.cpp feature matrix](https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix)
296
+
297
+ But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.
298
+
299
+ These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.
300
+
301
+ The I-quants are *not* compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.
302
+
303
+ Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski