webmp3 commited on
Commit
2f8dc0b
·
verified ·
1 Parent(s): c14aeae

Publish complete validated GGUF release

Browse files
.gitattributes CHANGED
@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
36
  Sakura-Ripple-65B-Preview-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
37
  Sakura-Ripple-65B-Preview-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
38
  Sakura-Ripple-65B-Preview-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
 
 
36
  Sakura-Ripple-65B-Preview-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
37
  Sakura-Ripple-65B-Preview-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
38
  Sakura-Ripple-65B-Preview-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
39
+ Sakura-Ripple-65B-Preview-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -31,6 +31,7 @@ Original: **129.6 GB (BF16 GGUF, the source of these quantizations)**.
31
  | Q2_K | 23.7 GB | 82 % smaller |
32
  | Q3_K_M | 31.0 GB | 76 % smaller |
33
  | Q4_K_M | 38.8 GB | 70 % smaller |
 
34
 
35
  ## 2. Original Model / Attribution
36
 
@@ -45,6 +46,7 @@ Original: **129.6 GB (BF16 GGUF, the source of these quantizations)**.
45
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23.7 GB | Q2_K |
46
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31.0 GB | Q3_K_M |
47
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38.8 GB | Q4_K_M |
 
48
 
49
 
50
 
@@ -79,6 +81,7 @@ Needs a llama.cpp build that supports this architecture. If a quantization is sp
79
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23747659040 | `60b79fc65989267fcde15035ed01bb64a3915864e6e30d85159c5c1db066cc20` | Q2_K |
80
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31019943200 | `fdf6815428dd2ed8ebf5d542525ca337bd1057d10d94bb74561736e24831574c` | Q3_K_M |
81
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38798820640 | `7f5a149ac052e8125745d4f4b369c5e82c7ed8b43a5da1cd2b2a68fd6440f197` | Q4_K_M |
 
82
 
83
  ### How these were made
84
 
@@ -93,6 +96,7 @@ The quant label names the llama.cpp preset. Stock llama.cpp keeps some tensors (
93
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | F32: 561, Q2_K: 401, Q3_K: 160, Q4_K: 80, Q6_K: 1 |
94
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | F32: 561, Q3_K: 401, Q4_K: 233, Q5_K: 7, Q6_K: 1 |
95
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | F32: 561, Q4_K: 561, Q6_K: 81 |
 
96
 
97
  ### Test builds
98
 
@@ -101,6 +105,7 @@ The quant label names the llama.cpp preset. Stock llama.cpp keeps some tensors (
101
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
102
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
103
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
 
104
 
105
  Q5_K_M will be added to this repository when it is finished and checked.
106
 
@@ -129,6 +134,7 @@ Source license: `apache-2.0`. apache-2.0 allows redistribution with attribution
129
  | Q2_K | 23.7 GB | 小 82 % |
130
  | Q3_K_M | 31.0 GB | 小 76 % |
131
  | Q4_K_M | 38.8 GB | 小 70 % |
 
132
 
133
  ## 2. 原始模型与署名
134
 
@@ -143,6 +149,7 @@ Source license: `apache-2.0`. apache-2.0 allows redistribution with attribution
143
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23.7 GB | Q2_K |
144
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31.0 GB | Q3_K_M |
145
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38.8 GB | Q4_K_M |
 
146
 
147
  ## 4. 下载与使用
148
 
@@ -175,6 +182,7 @@ llama-cli -m <file>.gguf -p "Hello" -n 32
175
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23747659040 | `60b79fc65989267fcde15035ed01bb64a3915864e6e30d85159c5c1db066cc20` | Q2_K |
176
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31019943200 | `fdf6815428dd2ed8ebf5d542525ca337bd1057d10d94bb74561736e24831574c` | Q3_K_M |
177
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38798820640 | `7f5a149ac052e8125745d4f4b369c5e82c7ed8b43a5da1cd2b2a68fd6440f197` | Q4_K_M |
 
178
 
179
  ### 制作方式
180
 
@@ -189,6 +197,7 @@ llama-cli -m <file>.gguf -p "Hello" -n 32
189
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | F32: 561, Q2_K: 401, Q3_K: 160, Q4_K: 80, Q6_K: 1 |
190
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | F32: 561, Q3_K: 401, Q4_K: 233, Q5_K: 7, Q6_K: 1 |
191
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | F32: 561, Q4_K: 561, Q6_K: 81 |
 
192
 
193
  ### 测试版本
194
 
@@ -197,6 +206,7 @@ llama-cli -m <file>.gguf -p "Hello" -n 32
197
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
198
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
199
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
 
200
 
201
  Q5_K_M 完成并通过检查后将添加到本仓库。
202
 
 
31
  | Q2_K | 23.7 GB | 82 % smaller |
32
  | Q3_K_M | 31.0 GB | 76 % smaller |
33
  | Q4_K_M | 38.8 GB | 70 % smaller |
34
+ | Q5_K_M | 45.8 GB | 65 % smaller |
35
 
36
  ## 2. Original Model / Attribution
37
 
 
46
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23.7 GB | Q2_K |
47
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31.0 GB | Q3_K_M |
48
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38.8 GB | Q4_K_M |
49
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | 45.8 GB | Q5_K_M |
50
 
51
 
52
 
 
81
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23747659040 | `60b79fc65989267fcde15035ed01bb64a3915864e6e30d85159c5c1db066cc20` | Q2_K |
82
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31019943200 | `fdf6815428dd2ed8ebf5d542525ca337bd1057d10d94bb74561736e24831574c` | Q3_K_M |
83
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38798820640 | `7f5a149ac052e8125745d4f4b369c5e82c7ed8b43a5da1cd2b2a68fd6440f197` | Q4_K_M |
84
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | 45757433120 | `a606978bedce45a40627773c7bd442d3c9d53c673e5c268d354bbb639950ffe4` | Q5_K_M |
85
 
86
  ### How these were made
87
 
 
96
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | F32: 561, Q2_K: 401, Q3_K: 160, Q4_K: 80, Q6_K: 1 |
97
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | F32: 561, Q3_K: 401, Q4_K: 233, Q5_K: 7, Q6_K: 1 |
98
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | F32: 561, Q4_K: 561, Q6_K: 81 |
99
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | F32: 561, Q5_K: 561, Q6_K: 81 |
100
 
101
  ### Test builds
102
 
 
105
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
106
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
107
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
108
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
109
 
110
  Q5_K_M will be added to this repository when it is finished and checked.
111
 
 
134
  | Q2_K | 23.7 GB | 小 82 % |
135
  | Q3_K_M | 31.0 GB | 小 76 % |
136
  | Q4_K_M | 38.8 GB | 小 70 % |
137
+ | Q5_K_M | 45.8 GB | 小 65 % |
138
 
139
  ## 2. 原始模型与署名
140
 
 
149
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23.7 GB | Q2_K |
150
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31.0 GB | Q3_K_M |
151
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38.8 GB | Q4_K_M |
152
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | 45.8 GB | Q5_K_M |
153
 
154
  ## 4. 下载与使用
155
 
 
182
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23747659040 | `60b79fc65989267fcde15035ed01bb64a3915864e6e30d85159c5c1db066cc20` | Q2_K |
183
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31019943200 | `fdf6815428dd2ed8ebf5d542525ca337bd1057d10d94bb74561736e24831574c` | Q3_K_M |
184
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38798820640 | `7f5a149ac052e8125745d4f4b369c5e82c7ed8b43a5da1cd2b2a68fd6440f197` | Q4_K_M |
185
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | 45757433120 | `a606978bedce45a40627773c7bd442d3c9d53c673e5c268d354bbb639950ffe4` | Q5_K_M |
186
 
187
  ### 制作方式
188
 
 
197
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | F32: 561, Q2_K: 401, Q3_K: 160, Q4_K: 80, Q6_K: 1 |
198
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | F32: 561, Q3_K: 401, Q4_K: 233, Q5_K: 7, Q6_K: 1 |
199
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | F32: 561, Q4_K: 561, Q6_K: 81 |
200
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | F32: 561, Q5_K: 561, Q6_K: 81 |
201
 
202
  ### 测试版本
203
 
 
206
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
207
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
208
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
209
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
210
 
211
  Q5_K_M 完成并通过检查后将添加到本仓库。
212
 
README_zh.md CHANGED
@@ -19,6 +19,7 @@
19
  | Q2_K | 23.7 GB | 小 82 % |
20
  | Q3_K_M | 31.0 GB | 小 76 % |
21
  | Q4_K_M | 38.8 GB | 小 70 % |
 
22
 
23
  ## 2. 原始模型与署名
24
 
@@ -33,6 +34,7 @@
33
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23.7 GB | Q2_K |
34
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31.0 GB | Q3_K_M |
35
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38.8 GB | Q4_K_M |
 
36
 
37
  ## 4. 下载与使用
38
 
@@ -65,6 +67,7 @@ llama-cli -m <file>.gguf -p "Hello" -n 32
65
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23747659040 | `60b79fc65989267fcde15035ed01bb64a3915864e6e30d85159c5c1db066cc20` | Q2_K |
66
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31019943200 | `fdf6815428dd2ed8ebf5d542525ca337bd1057d10d94bb74561736e24831574c` | Q3_K_M |
67
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38798820640 | `7f5a149ac052e8125745d4f4b369c5e82c7ed8b43a5da1cd2b2a68fd6440f197` | Q4_K_M |
 
68
 
69
  ### 制作方式
70
 
@@ -79,6 +82,7 @@ llama-cli -m <file>.gguf -p "Hello" -n 32
79
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | F32: 561, Q2_K: 401, Q3_K: 160, Q4_K: 80, Q6_K: 1 |
80
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | F32: 561, Q3_K: 401, Q4_K: 233, Q5_K: 7, Q6_K: 1 |
81
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | F32: 561, Q4_K: 561, Q6_K: 81 |
 
82
 
83
  ### 测试版本
84
 
@@ -87,6 +91,7 @@ llama-cli -m <file>.gguf -p "Hello" -n 32
87
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
88
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
89
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
 
90
 
91
  Q5_K_M 完成并通过检查后将添加到本仓库。
92
 
 
19
  | Q2_K | 23.7 GB | 小 82 % |
20
  | Q3_K_M | 31.0 GB | 小 76 % |
21
  | Q4_K_M | 38.8 GB | 小 70 % |
22
+ | Q5_K_M | 45.8 GB | 小 65 % |
23
 
24
  ## 2. 原始模型与署名
25
 
 
34
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23.7 GB | Q2_K |
35
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31.0 GB | Q3_K_M |
36
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38.8 GB | Q4_K_M |
37
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | 45.8 GB | Q5_K_M |
38
 
39
  ## 4. 下载与使用
40
 
 
67
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | 23747659040 | `60b79fc65989267fcde15035ed01bb64a3915864e6e30d85159c5c1db066cc20` | Q2_K |
68
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | 31019943200 | `fdf6815428dd2ed8ebf5d542525ca337bd1057d10d94bb74561736e24831574c` | Q3_K_M |
69
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | 38798820640 | `7f5a149ac052e8125745d4f4b369c5e82c7ed8b43a5da1cd2b2a68fd6440f197` | Q4_K_M |
70
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | 45757433120 | `a606978bedce45a40627773c7bd442d3c9d53c673e5c268d354bbb639950ffe4` | Q5_K_M |
71
 
72
  ### 制作方式
73
 
 
82
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | F32: 561, Q2_K: 401, Q3_K: 160, Q4_K: 80, Q6_K: 1 |
83
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | F32: 561, Q3_K: 401, Q4_K: 233, Q5_K: 7, Q6_K: 1 |
84
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | F32: 561, Q4_K: 561, Q6_K: 81 |
85
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | F32: 561, Q5_K: 561, Q6_K: 81 |
86
 
87
  ### 测试版本
88
 
 
91
  | `Sakura-Ripple-65B-Preview-Q2_K.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
92
  | `Sakura-Ripple-65B-Preview-Q3_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
93
  | `Sakura-Ripple-65B-Preview-Q4_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
94
+ | `Sakura-Ripple-65B-Preview-Q5_K_M.gguf` | `b12189-80d2c0c2c` | `d6cc36e5fdc2eedd6e052dc7eb1e29d21065331e37e31f57ab2ca4720635dba2` |
95
 
96
  Q5_K_M 完成并通过检查后将添加到本仓库。
97
 
Sakura-Ripple-65B-Preview-Q5_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a606978bedce45a40627773c7bd442d3c9d53c673e5c268d354bbb639950ffe4
3
+ size 45757433120