node-llama-cpp 3.0.3 → 3.1.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/bindings/Llama.d.ts +3 -0
- package/dist/bindings/Llama.js +7 -0
- package/dist/bindings/Llama.js.map +1 -1
- package/dist/chatWrappers/Llama3_1ChatWrapper.d.ts +1 -1
- package/dist/chatWrappers/Llama3_1ChatWrapper.js +2 -1
- package/dist/chatWrappers/Llama3_1ChatWrapper.js.map +1 -1
- package/dist/cli/commands/ChatCommand.js +2 -2
- package/dist/cli/commands/ChatCommand.js.map +1 -1
- package/dist/cli/commands/CompleteCommand.js +2 -2
- package/dist/cli/commands/CompleteCommand.js.map +1 -1
- package/dist/cli/commands/InfillCommand.js +2 -2
- package/dist/cli/commands/InfillCommand.js.map +1 -1
- package/dist/cli/commands/InitCommand.js +36 -20
- package/dist/cli/commands/InitCommand.js.map +1 -1
- package/dist/cli/commands/PullCommand.js +10 -11
- package/dist/cli/commands/PullCommand.js.map +1 -1
- package/dist/cli/commands/inspect/commands/InspectEstimateCommand.js +13 -16
- package/dist/cli/commands/inspect/commands/InspectEstimateCommand.js.map +1 -1
- package/dist/cli/commands/inspect/commands/InspectGgufCommand.js +18 -19
- package/dist/cli/commands/inspect/commands/InspectGgufCommand.js.map +1 -1
- package/dist/cli/commands/inspect/commands/InspectMeasureCommand.js +2 -2
- package/dist/cli/commands/inspect/commands/InspectMeasureCommand.js.map +1 -1
- package/dist/cli/recommendedModels.js +102 -403
- package/dist/cli/recommendedModels.js.map +1 -1
- package/dist/cli/utils/interactivelyAskForModel.js +49 -36
- package/dist/cli/utils/interactivelyAskForModel.js.map +1 -1
- package/dist/cli/utils/printModelDestination.d.ts +2 -0
- package/dist/cli/utils/printModelDestination.js +11 -0
- package/dist/cli/utils/printModelDestination.js.map +1 -0
- package/dist/cli/utils/projectTemplates.d.ts +2 -2
- package/dist/cli/utils/projectTemplates.js +2 -2
- package/dist/cli/utils/projectTemplates.js.map +1 -1
- package/dist/cli/utils/resolveCommandGgufPath.js +12 -14
- package/dist/cli/utils/resolveCommandGgufPath.js.map +1 -1
- package/dist/cli/utils/resolveModelRecommendationFileOptions.d.ts +2 -7
- package/dist/cli/utils/resolveModelRecommendationFileOptions.js +7 -2
- package/dist/cli/utils/resolveModelRecommendationFileOptions.js.map +1 -1
- package/dist/gguf/consts.js +1 -1
- package/dist/gguf/consts.js.map +1 -1
- package/dist/gguf/insights/GgufInsightsConfigurationResolver.d.ts +7 -2
- package/dist/gguf/insights/GgufInsightsConfigurationResolver.js +7 -3
- package/dist/gguf/insights/GgufInsightsConfigurationResolver.js.map +1 -1
- package/dist/gguf/readGgufFileInfo.d.ts +2 -2
- package/dist/gguf/readGgufFileInfo.js +12 -10
- package/dist/gguf/readGgufFileInfo.js.map +1 -1
- package/dist/gguf/utils/resolveBinarySplitGgufPartUrls.js +1 -1
- package/dist/gguf/utils/resolveBinarySplitGgufPartUrls.js.map +1 -1
- package/dist/gguf/utils/resolveSplitGgufParts.d.ts +1 -1
- package/dist/gguf/utils/resolveSplitGgufParts.js +15 -6
- package/dist/gguf/utils/resolveSplitGgufParts.js.map +1 -1
- package/dist/index.d.ts +3 -1
- package/dist/index.js +2 -1
- package/dist/index.js.map +1 -1
- package/dist/tsconfig.tsbuildinfo +1 -1
- package/dist/utils/createModelDownloader.d.ts +45 -8
- package/dist/utils/createModelDownloader.js +43 -13
- package/dist/utils/createModelDownloader.js.map +1 -1
- package/dist/utils/getLlamaClasses.d.ts +9 -0
- package/dist/utils/getLlamaClasses.js +14 -0
- package/dist/utils/getLlamaClasses.js.map +1 -0
- package/dist/utils/isUrl.js +1 -1
- package/dist/utils/isUrl.js.map +1 -1
- package/dist/utils/modelFileAccesTokens.js +2 -1
- package/dist/utils/modelFileAccesTokens.js.map +1 -1
- package/dist/utils/parseModelUri.d.ts +9 -0
- package/dist/utils/parseModelUri.js +128 -0
- package/dist/utils/parseModelUri.js.map +1 -0
- package/dist/utils/resolveModelDestination.d.ts +14 -0
- package/dist/utils/resolveModelDestination.js +31 -0
- package/dist/utils/resolveModelDestination.js.map +1 -0
- package/dist/utils/resolveModelFile.d.ts +136 -0
- package/dist/utils/resolveModelFile.js +148 -0
- package/dist/utils/resolveModelFile.js.map +1 -0
- package/llama/binariesGithubRelease.json +1 -1
- package/llama/gitRelease.bundle +0 -0
- package/llama/llama.cpp.info.json +1 -1
- package/package.json +12 -12
- package/templates/packed/electron-typescript-react.json +1 -1
- package/templates/packed/node-typescript.json +1 -1
|
@@ -3,124 +3,54 @@ export const recommendedModels = [{
|
|
|
3
3
|
abilities: ["chat", "complete", "functionCalling"],
|
|
4
4
|
description: "Llama 3.1 model was created by Meta and is optimized for an assistant-like chat use cases, with support for function calling.\n" +
|
|
5
5
|
"This is the 8 billion parameters version of the model.",
|
|
6
|
-
fileOptions: [
|
|
7
|
-
|
|
8
|
-
|
|
9
|
-
|
|
10
|
-
|
|
11
|
-
}
|
|
12
|
-
}, {
|
|
13
|
-
huggingFace: {
|
|
14
|
-
model: "mradermacher/Meta-Llama-3.1-8B-Instruct-GGUF",
|
|
15
|
-
branch: "main",
|
|
16
|
-
file: "Meta-Llama-3.1-8B-Instruct.Q6_K.gguf"
|
|
17
|
-
}
|
|
18
|
-
}, {
|
|
19
|
-
huggingFace: {
|
|
20
|
-
model: "mradermacher/Meta-Llama-3.1-8B-Instruct-GGUF",
|
|
21
|
-
branch: "main",
|
|
22
|
-
file: "Meta-Llama-3.1-8B-Instruct.Q4_K_M.gguf"
|
|
23
|
-
}
|
|
24
|
-
}]
|
|
6
|
+
fileOptions: [
|
|
7
|
+
"hf:mradermacher/Meta-Llama-3.1-8B-Instruct-GGUF/Meta-Llama-3.1-8B-Instruct.Q8_0.gguf",
|
|
8
|
+
"hf:mradermacher/Meta-Llama-3.1-8B-Instruct-GGUF/Meta-Llama-3.1-8B-Instruct.Q6_K.gguf",
|
|
9
|
+
"hf:mradermacher/Meta-Llama-3.1-8B-Instruct-GGUF/Meta-Llama-3.1-8B-Instruct.Q4_K_M.gguf"
|
|
10
|
+
]
|
|
25
11
|
}, {
|
|
26
12
|
name: "Llama 3.1 70B",
|
|
27
13
|
abilities: ["chat", "complete", "functionCalling"],
|
|
28
14
|
description: "Llama 3.1 model was created by Meta and is optimized for an assistant-like chat use cases, with support for function calling.\n" +
|
|
29
15
|
"This is the 70 billion parameters version of the model. " +
|
|
30
16
|
"You need a GPU with a lot of VRAM to use this version.",
|
|
31
|
-
fileOptions: [
|
|
32
|
-
|
|
33
|
-
|
|
34
|
-
|
|
35
|
-
|
|
36
|
-
|
|
37
|
-
}, {
|
|
38
|
-
huggingFace: {
|
|
39
|
-
model: "mradermacher/Meta-Llama-3.1-70B-Instruct-GGUF",
|
|
40
|
-
branch: "main",
|
|
41
|
-
file: "Meta-Llama-3.1-70B-Instruct.Q6_K.gguf.part1of2"
|
|
42
|
-
}
|
|
43
|
-
}, {
|
|
44
|
-
huggingFace: {
|
|
45
|
-
model: "mradermacher/Meta-Llama-3.1-70B-Instruct-GGUF",
|
|
46
|
-
branch: "main",
|
|
47
|
-
file: "Meta-Llama-3.1-70B-Instruct.Q4_K_M.gguf"
|
|
48
|
-
}
|
|
49
|
-
}, {
|
|
50
|
-
huggingFace: {
|
|
51
|
-
model: "mradermacher/Meta-Llama-3.1-70B-Instruct-GGUF",
|
|
52
|
-
branch: "main",
|
|
53
|
-
file: "Meta-Llama-3.1-70B-Instruct.Q4_K_S.gguf"
|
|
54
|
-
}
|
|
55
|
-
}]
|
|
17
|
+
fileOptions: [
|
|
18
|
+
"hf:mradermacher/Meta-Llama-3.1-70B-Instruct-GGUF/Meta-Llama-3.1-70B-Instruct.Q8_0.gguf.part1of2",
|
|
19
|
+
"hf:mradermacher/Meta-Llama-3.1-70B-Instruct-GGUF/Meta-Llama-3.1-70B-Instruct.Q6_K.gguf.part1of2",
|
|
20
|
+
"hf:mradermacher/Meta-Llama-3.1-70B-Instruct-GGUF/Meta-Llama-3.1-70B-Instruct.Q4_K_M.gguf",
|
|
21
|
+
"hf:mradermacher/Meta-Llama-3.1-70B-Instruct-GGUF/Meta-Llama-3.1-70B-Instruct.Q4_K_S.gguf"
|
|
22
|
+
]
|
|
56
23
|
}, {
|
|
57
24
|
name: "Llama 3.1 405B",
|
|
58
25
|
abilities: ["chat", "complete", "functionCalling"],
|
|
59
26
|
description: "Llama 3.1 model was created by Meta and is optimized for an assistant-like chat use cases, with support for function calling.\n" +
|
|
60
27
|
"This is the 405 billion parameters version of the model, and its capabilities are comparable and sometimes even surpass GPT-4o and Claude 3.5 Sonnet.\n" +
|
|
61
28
|
"You need a GPU with a lot of VRAM to use this version of Llama 3.1.",
|
|
62
|
-
fileOptions: [
|
|
63
|
-
|
|
64
|
-
|
|
65
|
-
|
|
66
|
-
file: "Meta-Llama-3.1-405B-Instruct.Q3_K_L.gguf.part1of5"
|
|
67
|
-
}
|
|
68
|
-
}, {
|
|
69
|
-
huggingFace: {
|
|
70
|
-
model: "mradermacher/Meta-Llama-3.1-405B-Instruct-GGUF",
|
|
71
|
-
branch: "main",
|
|
72
|
-
file: "Meta-Llama-3.1-405B-Instruct.Q3_K_M.gguf.part1of4"
|
|
73
|
-
}
|
|
74
|
-
}]
|
|
29
|
+
fileOptions: [
|
|
30
|
+
"hf:mradermacher/Meta-Llama-3.1-405B-Instruct-GGUF/Meta-Llama-3.1-405B-Instruct.Q3_K_L.gguf.part1of5",
|
|
31
|
+
"hf:mradermacher/Meta-Llama-3.1-405B-Instruct-GGUF/Meta-Llama-3.1-405B-Instruct.Q3_K_M.gguf.part1of4"
|
|
32
|
+
]
|
|
75
33
|
}, {
|
|
76
34
|
name: "Mistral Nemo 12B",
|
|
77
35
|
abilities: ["chat", "complete", "functionCalling"],
|
|
78
36
|
description: "Mistral Nemo model was created by Mistral AI and was trained on large proportion of multilingual and code data, with support for function calling.\n" +
|
|
79
37
|
"It was trained jointly by Mistral AI and NVIDIA.\n" +
|
|
80
38
|
"This is a 12 billion parameters model.",
|
|
81
|
-
fileOptions: [
|
|
82
|
-
|
|
83
|
-
|
|
84
|
-
|
|
85
|
-
|
|
86
|
-
|
|
87
|
-
}, {
|
|
88
|
-
huggingFace: {
|
|
89
|
-
model: "mradermacher/Mistral-Nemo-Instruct-2407-GGUF",
|
|
90
|
-
branch: "main",
|
|
91
|
-
file: "Mistral-Nemo-Instruct-2407.Q6_K.gguf"
|
|
92
|
-
}
|
|
93
|
-
}, {
|
|
94
|
-
huggingFace: {
|
|
95
|
-
model: "mradermacher/Mistral-Nemo-Instruct-2407-GGUF",
|
|
96
|
-
branch: "main",
|
|
97
|
-
file: "Mistral-Nemo-Instruct-2407.Q4_K_M.gguf"
|
|
98
|
-
}
|
|
99
|
-
}, {
|
|
100
|
-
huggingFace: {
|
|
101
|
-
model: "mradermacher/Mistral-Nemo-Instruct-2407-GGUF",
|
|
102
|
-
branch: "main",
|
|
103
|
-
file: "Mistral-Nemo-Instruct-2407.Q4_K_S.gguf"
|
|
104
|
-
}
|
|
105
|
-
}]
|
|
39
|
+
fileOptions: [
|
|
40
|
+
"hf:mradermacher/Mistral-Nemo-Instruct-2407-GGUF/Mistral-Nemo-Instruct-2407.Q8_0.gguf",
|
|
41
|
+
"hf:mradermacher/Mistral-Nemo-Instruct-2407-GGUF/Mistral-Nemo-Instruct-2407.Q6_K.gguf",
|
|
42
|
+
"hf:mradermacher/Mistral-Nemo-Instruct-2407-GGUF/Mistral-Nemo-Instruct-2407.Q4_K_M.gguf",
|
|
43
|
+
"hf:mradermacher/Mistral-Nemo-Instruct-2407-GGUF/Mistral-Nemo-Instruct-2407.Q4_K_S.gguf"
|
|
44
|
+
]
|
|
106
45
|
}, {
|
|
107
46
|
name: "Phi 3 3.8B",
|
|
108
47
|
abilities: ["chat", "complete", "functionCalling"],
|
|
109
48
|
description: "Phi 3 model was created by Microsoft and is optimized for strong reasoning (especially math and logic).\n" +
|
|
110
49
|
"This is the smallversion of the model.",
|
|
111
|
-
fileOptions: [
|
|
112
|
-
|
|
113
|
-
|
|
114
|
-
|
|
115
|
-
file: "Phi-3.1-mini-4k-instruct-Q8_0.gguf"
|
|
116
|
-
}
|
|
117
|
-
}, {
|
|
118
|
-
huggingFace: {
|
|
119
|
-
model: "bartowski/Phi-3.1-mini-4k-instruct-GGUF",
|
|
120
|
-
branch: "main",
|
|
121
|
-
file: "Phi-3.1-mini-4k-instruct-Q4_K_M.gguf"
|
|
122
|
-
}
|
|
123
|
-
}]
|
|
50
|
+
fileOptions: [
|
|
51
|
+
"hf:bartowski/Phi-3.1-mini-4k-instruct-GGUF/Phi-3.1-mini-4k-instruct-Q8_0.gguf",
|
|
52
|
+
"hf:bartowski/Phi-3.1-mini-4k-instruct-GGUF/Phi-3.1-mini-4k-instruct-Q4_K_M.gguf"
|
|
53
|
+
]
|
|
124
54
|
}, {
|
|
125
55
|
name: "OLMoE 1B 7B MoE",
|
|
126
56
|
abilities: ["chat"],
|
|
@@ -128,179 +58,70 @@ export const recommendedModels = [{
|
|
|
128
58
|
"Mixtures of Experts (MoE) is a technique where different models, each skilled in solving a particular kind of problem, work together to the improve the overall performance on complex tasks.\n" +
|
|
129
59
|
"This model includes 64 expert models, with a total of 7 billion parameters.\n" +
|
|
130
60
|
"This model generates output extremely fast.",
|
|
131
|
-
fileOptions: [
|
|
132
|
-
|
|
133
|
-
|
|
134
|
-
|
|
135
|
-
|
|
136
|
-
|
|
137
|
-
|
|
138
|
-
huggingFace: {
|
|
139
|
-
model: "allenai/OLMoE-1B-7B-0924-Instruct-GGUF",
|
|
140
|
-
branch: "main",
|
|
141
|
-
file: "olmoe-1b-7b-0924-instruct-q6_k.gguf"
|
|
142
|
-
}
|
|
143
|
-
}, {
|
|
144
|
-
huggingFace: {
|
|
145
|
-
model: "allenai/OLMoE-1B-7B-0924-Instruct-GGUF",
|
|
146
|
-
branch: "main",
|
|
147
|
-
file: "olmoe-1b-7b-0924-instruct-q5_k_m.gguf"
|
|
148
|
-
}
|
|
149
|
-
}, {
|
|
150
|
-
huggingFace: {
|
|
151
|
-
model: "allenai/OLMoE-1B-7B-0924-Instruct-GGUF",
|
|
152
|
-
branch: "main",
|
|
153
|
-
file: "olmoe-1b-7b-0924-instruct-q4_k_s.gguf"
|
|
154
|
-
}
|
|
155
|
-
}, {
|
|
156
|
-
huggingFace: {
|
|
157
|
-
model: "allenai/OLMoE-1B-7B-0924-Instruct-GGUF",
|
|
158
|
-
branch: "main",
|
|
159
|
-
file: "olmoe-1b-7b-0924-instruct-q4_k_m.gguf"
|
|
160
|
-
}
|
|
161
|
-
}]
|
|
61
|
+
fileOptions: [
|
|
62
|
+
"hf:allenai/OLMoE-1B-7B-0924-Instruct-GGUF/olmoe-1b-7b-0924-instruct-q8_0.gguf",
|
|
63
|
+
"hf:allenai/OLMoE-1B-7B-0924-Instruct-GGUF/olmoe-1b-7b-0924-instruct-q6_k.gguf",
|
|
64
|
+
"hf:allenai/OLMoE-1B-7B-0924-Instruct-GGUF/olmoe-1b-7b-0924-instruct-q5_k_m.gguf",
|
|
65
|
+
"hf:allenai/OLMoE-1B-7B-0924-Instruct-GGUF/olmoe-1b-7b-0924-instruct-q4_k_s.gguf",
|
|
66
|
+
"hf:allenai/OLMoE-1B-7B-0924-Instruct-GGUF/olmoe-1b-7b-0924-instruct-q4_k_m.gguf"
|
|
67
|
+
]
|
|
162
68
|
}, {
|
|
163
69
|
name: "Mixtral 8x7B MoE",
|
|
164
70
|
abilities: ["chat", "complete"],
|
|
165
71
|
description: "Mixtral models were created by Mistal AI and are general purpose models that utilize a Mixture of Experts architecture.\n" +
|
|
166
72
|
"Mixtures of Experts (MoE) is a technique where different models, each skilled in solving a particular kind of problem, work together to the improve the overall performance on complex tasks.\n" +
|
|
167
73
|
"This model includes 8 expert models, each with 7 billion parameters.",
|
|
168
|
-
fileOptions: [
|
|
169
|
-
|
|
170
|
-
|
|
171
|
-
|
|
172
|
-
file: "mixtral-8x7b-v0.1.Q5_K_M.gguf"
|
|
173
|
-
}
|
|
174
|
-
}, {
|
|
175
|
-
huggingFace: {
|
|
176
|
-
model: "TheBloke/Mixtral-8x7B-v0.1-GGUF",
|
|
177
|
-
branch: "main",
|
|
178
|
-
file: "mixtral-8x7b-v0.1.Q4_K_M.gguf"
|
|
179
|
-
}
|
|
180
|
-
}]
|
|
74
|
+
fileOptions: [
|
|
75
|
+
"hf:TheBloke/Mixtral-8x7B-v0.1-GGUF/mixtral-8x7b-v0.1.Q5_K_M.gguf",
|
|
76
|
+
"hf:TheBloke/Mixtral-8x7B-v0.1-GGUF/mixtral-8x7b-v0.1.Q4_K_M.gguf"
|
|
77
|
+
]
|
|
181
78
|
}, {
|
|
182
79
|
name: "Mistral 7B Instruct v0.2",
|
|
183
80
|
abilities: ["chat", "complete"],
|
|
184
81
|
description: "Mistral models were created by Mistal AI and are general purpose models.\n" +
|
|
185
82
|
"This is the 7 billion parameters version of the model.",
|
|
186
|
-
fileOptions: [
|
|
187
|
-
|
|
188
|
-
|
|
189
|
-
|
|
190
|
-
file: "mistral-7b-instruct-v0.2.Q5_K_M.gguf"
|
|
191
|
-
}
|
|
192
|
-
}, {
|
|
193
|
-
huggingFace: {
|
|
194
|
-
model: "TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
|
|
195
|
-
branch: "main",
|
|
196
|
-
file: "mistral-7b-instruct-v0.2.Q4_K_M.gguf"
|
|
197
|
-
}
|
|
198
|
-
}]
|
|
83
|
+
fileOptions: [
|
|
84
|
+
"hf:TheBloke/Mistral-7B-Instruct-v0.2-GGUF/mistral-7b-instruct-v0.2.Q5_K_M.gguf",
|
|
85
|
+
"hf:TheBloke/Mistral-7B-Instruct-v0.2-GGUF/mistral-7b-instruct-v0.2.Q4_K_M.gguf"
|
|
86
|
+
]
|
|
199
87
|
}, {
|
|
200
88
|
name: "Dolphin 2.5 Mixtral 8x7B MoE",
|
|
201
89
|
abilities: ["chat", "complete"],
|
|
202
90
|
description: "This Dolphin Mixtral model was created by Eric Hartford and is an uncensored model based on Mixtral, with really good coding skills.\n" +
|
|
203
91
|
"See the Mixtral model above for more information about Mixtral models.\n" +
|
|
204
92
|
"This model includes 8 expert models, each with 7 billion parameters.",
|
|
205
|
-
fileOptions: [
|
|
206
|
-
|
|
207
|
-
|
|
208
|
-
|
|
209
|
-
file: "dolphin-2.5-mixtral-8x7b.Q5_K_M.gguf"
|
|
210
|
-
}
|
|
211
|
-
}, {
|
|
212
|
-
huggingFace: {
|
|
213
|
-
model: "TheBloke/dolphin-2.5-mixtral-8x7b-GGUF",
|
|
214
|
-
branch: "main",
|
|
215
|
-
file: "dolphin-2.5-mixtral-8x7b.Q4_K_M.gguf"
|
|
216
|
-
}
|
|
217
|
-
}]
|
|
93
|
+
fileOptions: [
|
|
94
|
+
"hf:TheBloke/dolphin-2.5-mixtral-8x7b-GGUF/dolphin-2.5-mixtral-8x7b.Q5_K_M.gguf",
|
|
95
|
+
"hf:TheBloke/dolphin-2.5-mixtral-8x7b-GGUF/dolphin-2.5-mixtral-8x7b.Q4_K_M.gguf"
|
|
96
|
+
]
|
|
218
97
|
}, {
|
|
219
98
|
name: "Gemma 2 9B",
|
|
220
99
|
abilities: ["chat", "complete"],
|
|
221
100
|
description: "Gemma models were created by Google and are optimized suited for variety of text generation tasks, " +
|
|
222
101
|
"including question answering, summarization, and reasoning, with a focus on responsible responses.\n" +
|
|
223
102
|
"This is the 9 billion parameters version of the model.",
|
|
224
|
-
fileOptions: [
|
|
225
|
-
|
|
226
|
-
|
|
227
|
-
|
|
228
|
-
|
|
229
|
-
|
|
230
|
-
|
|
231
|
-
|
|
232
|
-
|
|
233
|
-
branch: "main",
|
|
234
|
-
file: "gemma-2-9b-it-Q6_K.gguf"
|
|
235
|
-
}
|
|
236
|
-
}, {
|
|
237
|
-
huggingFace: {
|
|
238
|
-
model: "bartowski/gemma-2-9b-it-GGUF",
|
|
239
|
-
branch: "main",
|
|
240
|
-
file: "gemma-2-9b-it-Q5_K_L.gguf"
|
|
241
|
-
}
|
|
242
|
-
}, {
|
|
243
|
-
huggingFace: {
|
|
244
|
-
model: "bartowski/gemma-2-9b-it-GGUF",
|
|
245
|
-
branch: "main",
|
|
246
|
-
file: "gemma-2-9b-it-Q5_K_M.gguf"
|
|
247
|
-
}
|
|
248
|
-
}, {
|
|
249
|
-
huggingFace: {
|
|
250
|
-
model: "bartowski/gemma-2-9b-it-GGUF",
|
|
251
|
-
branch: "main",
|
|
252
|
-
file: "gemma-2-9b-it-Q5_K_S.gguf"
|
|
253
|
-
}
|
|
254
|
-
}, {
|
|
255
|
-
huggingFace: {
|
|
256
|
-
model: "bartowski/gemma-2-9b-it-GGUF",
|
|
257
|
-
branch: "main",
|
|
258
|
-
file: "gemma-2-9b-it-Q4_K_L.gguf"
|
|
259
|
-
}
|
|
260
|
-
}, {
|
|
261
|
-
huggingFace: {
|
|
262
|
-
model: "bartowski/gemma-2-9b-it-GGUF",
|
|
263
|
-
branch: "main",
|
|
264
|
-
file: "gemma-2-9b-it-Q4_K_M.gguf"
|
|
265
|
-
}
|
|
266
|
-
}]
|
|
103
|
+
fileOptions: [
|
|
104
|
+
"hf:bartowski/gemma-2-9b-it-GGUF/gemma-2-9b-it-Q6_K_L.gguf",
|
|
105
|
+
"hf:bartowski/gemma-2-9b-it-GGUF/gemma-2-9b-it-Q6_K.gguf",
|
|
106
|
+
"hf:bartowski/gemma-2-9b-it-GGUF/gemma-2-9b-it-Q5_K_L.gguf",
|
|
107
|
+
"hf:bartowski/gemma-2-9b-it-GGUF/gemma-2-9b-it-Q5_K_M.gguf",
|
|
108
|
+
"hf:bartowski/gemma-2-9b-it-GGUF/gemma-2-9b-it-Q5_K_S.gguf",
|
|
109
|
+
"hf:bartowski/gemma-2-9b-it-GGUF/gemma-2-9b-it-Q4_K_L.gguf",
|
|
110
|
+
"hf:bartowski/gemma-2-9b-it-GGUF/gemma-2-9b-it-Q4_K_M.gguf"
|
|
111
|
+
]
|
|
267
112
|
}, {
|
|
268
113
|
name: "Gemma 2 2B",
|
|
269
114
|
abilities: ["chat", "complete"],
|
|
270
115
|
description: "Gemma models were created by Google and are optimized suited for variety of text generation tasks, " +
|
|
271
116
|
"including question answering, summarization, and reasoning, with a focus on responsible responses.\n" +
|
|
272
117
|
"This is the 2 billion parameters version of the model and is significantly less powerful than the 9B version.",
|
|
273
|
-
fileOptions: [
|
|
274
|
-
|
|
275
|
-
|
|
276
|
-
|
|
277
|
-
|
|
278
|
-
|
|
279
|
-
|
|
280
|
-
huggingFace: {
|
|
281
|
-
model: "bartowski/gemma-2-2b-it-GGUF",
|
|
282
|
-
branch: "main",
|
|
283
|
-
file: "gemma-2-2b-it-Q6_K.gguf"
|
|
284
|
-
}
|
|
285
|
-
}, {
|
|
286
|
-
huggingFace: {
|
|
287
|
-
model: "bartowski/gemma-2-2b-it-GGUF",
|
|
288
|
-
branch: "main",
|
|
289
|
-
file: "gemma-2-2b-it-Q5_K_M.gguf"
|
|
290
|
-
}
|
|
291
|
-
}, {
|
|
292
|
-
huggingFace: {
|
|
293
|
-
model: "bartowski/gemma-2-2b-it-GGUF",
|
|
294
|
-
branch: "main",
|
|
295
|
-
file: "gemma-2-2b-it-Q5_K_S.gguf"
|
|
296
|
-
}
|
|
297
|
-
}, {
|
|
298
|
-
huggingFace: {
|
|
299
|
-
model: "bartowski/gemma-2-2b-it-GGUF",
|
|
300
|
-
branch: "main",
|
|
301
|
-
file: "gemma-2-2b-it-Q4_K_M.gguf"
|
|
302
|
-
}
|
|
303
|
-
}]
|
|
118
|
+
fileOptions: [
|
|
119
|
+
"hf:bartowski/gemma-2-2b-it-GGUF/gemma-2-2b-it-Q6_K_L.gguf",
|
|
120
|
+
"hf:bartowski/gemma-2-2b-it-GGUF/gemma-2-2b-it-Q6_K.gguf",
|
|
121
|
+
"hf:bartowski/gemma-2-2b-it-GGUF/gemma-2-2b-it-Q5_K_M.gguf",
|
|
122
|
+
"hf:bartowski/gemma-2-2b-it-GGUF/gemma-2-2b-it-Q5_K_S.gguf",
|
|
123
|
+
"hf:bartowski/gemma-2-2b-it-GGUF/gemma-2-2b-it-Q4_K_M.gguf"
|
|
124
|
+
]
|
|
304
125
|
}, {
|
|
305
126
|
name: "Gemma 2 27B",
|
|
306
127
|
abilities: ["chat", "complete"],
|
|
@@ -308,122 +129,52 @@ export const recommendedModels = [{
|
|
|
308
129
|
"including question answering, summarization, and reasoning, with a focus on responsible responses.\n" +
|
|
309
130
|
"This is the 27 billion parameters version of the model.\n" +
|
|
310
131
|
"Since the model is relatively big, it may not run well on your machine",
|
|
311
|
-
fileOptions: [
|
|
312
|
-
|
|
313
|
-
|
|
314
|
-
|
|
315
|
-
|
|
316
|
-
|
|
317
|
-
|
|
318
|
-
|
|
319
|
-
|
|
320
|
-
branch: "main",
|
|
321
|
-
file: "gemma-2-27b-it-Q6_K.gguf"
|
|
322
|
-
}
|
|
323
|
-
}, {
|
|
324
|
-
huggingFace: {
|
|
325
|
-
model: "bartowski/gemma-2-27b-it-GGUF",
|
|
326
|
-
branch: "main",
|
|
327
|
-
file: "gemma-2-27b-it-Q5_K_L.gguf"
|
|
328
|
-
}
|
|
329
|
-
}, {
|
|
330
|
-
huggingFace: {
|
|
331
|
-
model: "bartowski/gemma-2-27b-it-GGUF",
|
|
332
|
-
branch: "main",
|
|
333
|
-
file: "gemma-2-27b-it-Q5_K_M.gguf"
|
|
334
|
-
}
|
|
335
|
-
}, {
|
|
336
|
-
huggingFace: {
|
|
337
|
-
model: "bartowski/gemma-2-27b-it-GGUF",
|
|
338
|
-
branch: "main",
|
|
339
|
-
file: "gemma-2-27b-it-Q5_K_S.gguf"
|
|
340
|
-
}
|
|
341
|
-
}, {
|
|
342
|
-
huggingFace: {
|
|
343
|
-
model: "bartowski/gemma-2-27b-it-GGUF",
|
|
344
|
-
branch: "main",
|
|
345
|
-
file: "gemma-2-27b-it-Q4_K_L.gguf"
|
|
346
|
-
}
|
|
347
|
-
}, {
|
|
348
|
-
huggingFace: {
|
|
349
|
-
model: "bartowski/gemma-2-27b-it-GGUF",
|
|
350
|
-
branch: "main",
|
|
351
|
-
file: "gemma-2-27b-it-Q4_K_M.gguf"
|
|
352
|
-
}
|
|
353
|
-
}]
|
|
132
|
+
fileOptions: [
|
|
133
|
+
"hf:bartowski/gemma-2-27b-it-GGUF/gemma-2-27b-it-Q6_K_L.gguf",
|
|
134
|
+
"hf:bartowski/gemma-2-27b-it-GGUF/gemma-2-27b-it-Q6_K.gguf",
|
|
135
|
+
"hf:bartowski/gemma-2-27b-it-GGUF/gemma-2-27b-it-Q5_K_L.gguf",
|
|
136
|
+
"hf:bartowski/gemma-2-27b-it-GGUF/gemma-2-27b-it-Q5_K_M.gguf",
|
|
137
|
+
"hf:bartowski/gemma-2-27b-it-GGUF/gemma-2-27b-it-Q5_K_S.gguf",
|
|
138
|
+
"hf:bartowski/gemma-2-27b-it-GGUF/gemma-2-27b-it-Q4_K_L.gguf",
|
|
139
|
+
"hf:bartowski/gemma-2-27b-it-GGUF/gemma-2-27b-it-Q4_K_M.gguf"
|
|
140
|
+
]
|
|
354
141
|
}, {
|
|
355
142
|
name: "Orca 2 13B",
|
|
356
143
|
abilities: ["chat", "complete"],
|
|
357
144
|
description: "Orca 2 model was created by Microsoft and is optimized for reasoning over given data, reading comprehensions, math problem solving and text summarization.\n" +
|
|
358
145
|
"This is the 13 billion parameters version of the model.",
|
|
359
|
-
fileOptions: [
|
|
360
|
-
|
|
361
|
-
|
|
362
|
-
|
|
363
|
-
file: "orca-2-13b.Q5_K_M.gguf"
|
|
364
|
-
}
|
|
365
|
-
}, {
|
|
366
|
-
huggingFace: {
|
|
367
|
-
model: "TheBloke/Orca-2-13B-GGUF",
|
|
368
|
-
branch: "main",
|
|
369
|
-
file: "orca-2-13b.Q4_K_M.gguf"
|
|
370
|
-
}
|
|
371
|
-
}]
|
|
146
|
+
fileOptions: [
|
|
147
|
+
"hf:TheBloke/Orca-2-13B-GGUF/orca-2-13b.Q5_K_M.gguf",
|
|
148
|
+
"hf:TheBloke/Orca-2-13B-GGUF/orca-2-13b.Q4_K_M.gguf"
|
|
149
|
+
]
|
|
372
150
|
}, {
|
|
373
151
|
name: "Code Llama 7B",
|
|
374
152
|
abilities: ["chat", "complete", "infill"],
|
|
375
153
|
description: "Code Llama model was created by Meta based on Llama 2 and is optimized for coding tasks.\n" +
|
|
376
154
|
"This is the 7 billion parameters version of the model.",
|
|
377
|
-
fileOptions: [
|
|
378
|
-
|
|
379
|
-
|
|
380
|
-
|
|
381
|
-
file: "codellama-7b.Q5_K_M.gguf"
|
|
382
|
-
}
|
|
383
|
-
}, {
|
|
384
|
-
huggingFace: {
|
|
385
|
-
model: "TheBloke/CodeLlama-7B-GGUF",
|
|
386
|
-
branch: "main",
|
|
387
|
-
file: "codellama-7b.Q4_K_M.gguf"
|
|
388
|
-
}
|
|
389
|
-
}]
|
|
155
|
+
fileOptions: [
|
|
156
|
+
"hf:TheBloke/CodeLlama-7B-GGUF/codellama-7b.Q5_K_M.gguf",
|
|
157
|
+
"hf:TheBloke/CodeLlama-7B-GGUF/codellama-7b.Q4_K_M.gguf"
|
|
158
|
+
]
|
|
390
159
|
}, {
|
|
391
160
|
name: "Code Llama 13B",
|
|
392
161
|
abilities: ["chat", "complete", "infill"],
|
|
393
162
|
description: "Code Llama model was created by Meta based on Llama 2 and is optimized for coding tasks.\n" +
|
|
394
163
|
"This is the 13 billion parameters version of the model.",
|
|
395
|
-
fileOptions: [
|
|
396
|
-
|
|
397
|
-
|
|
398
|
-
|
|
399
|
-
file: "codellama-13b.Q5_K_M.gguf"
|
|
400
|
-
}
|
|
401
|
-
}, {
|
|
402
|
-
huggingFace: {
|
|
403
|
-
model: "TheBloke/CodeLlama-13B-GGUF",
|
|
404
|
-
branch: "main",
|
|
405
|
-
file: "codellama-13b.Q4_K_M.gguf"
|
|
406
|
-
}
|
|
407
|
-
}]
|
|
164
|
+
fileOptions: [
|
|
165
|
+
"hf:TheBloke/CodeLlama-13B-GGUF/codellama-13b.Q5_K_M.gguf",
|
|
166
|
+
"hf:TheBloke/CodeLlama-13B-GGUF/codellama-13b.Q4_K_M.gguf"
|
|
167
|
+
]
|
|
408
168
|
}, {
|
|
409
169
|
name: "Code Llama 34B",
|
|
410
170
|
abilities: ["chat", "complete", "infill"],
|
|
411
171
|
description: "Code Llama model was created by Meta based on Llama 2 and is optimized for coding tasks.\n" +
|
|
412
172
|
"This is the 34 billion parameters version of the model.\n" +
|
|
413
173
|
"You need a GPU with handful of VRAM to use this version.",
|
|
414
|
-
fileOptions: [
|
|
415
|
-
|
|
416
|
-
|
|
417
|
-
|
|
418
|
-
file: "codellama-34b.Q5_K_M.gguf"
|
|
419
|
-
}
|
|
420
|
-
}, {
|
|
421
|
-
huggingFace: {
|
|
422
|
-
model: "TheBloke/CodeLlama-34B-GGUF",
|
|
423
|
-
branch: "main",
|
|
424
|
-
file: "codellama-34b.Q4_K_M.gguf"
|
|
425
|
-
}
|
|
426
|
-
}]
|
|
174
|
+
fileOptions: [
|
|
175
|
+
"hf:TheBloke/CodeLlama-34B-GGUF/codellama-34b.Q5_K_M.gguf",
|
|
176
|
+
"hf:TheBloke/CodeLlama-34B-GGUF/codellama-34b.Q4_K_M.gguf"
|
|
177
|
+
]
|
|
427
178
|
}, {
|
|
428
179
|
name: "CodeGemma 2B",
|
|
429
180
|
abilities: ["code", "complete", "infill"],
|
|
@@ -431,37 +182,13 @@ export const recommendedModels = [{
|
|
|
431
182
|
"natual language understanding, mathematical reasoning, and instruction following.\n" +
|
|
432
183
|
"This model is not suited for chat.\n" +
|
|
433
184
|
"This is the 2 billion parameters version of the model.\n",
|
|
434
|
-
fileOptions: [
|
|
435
|
-
|
|
436
|
-
|
|
437
|
-
|
|
438
|
-
|
|
439
|
-
|
|
440
|
-
|
|
441
|
-
huggingFace: {
|
|
442
|
-
model: "bartowski/codegemma-2b-GGUF",
|
|
443
|
-
branch: "main",
|
|
444
|
-
file: "codegemma-2b-Q6_K.gguf"
|
|
445
|
-
}
|
|
446
|
-
}, {
|
|
447
|
-
huggingFace: {
|
|
448
|
-
model: "bartowski/codegemma-2b-GGUF",
|
|
449
|
-
branch: "main",
|
|
450
|
-
file: "codegemma-2b-Q5_K_M.gguf"
|
|
451
|
-
}
|
|
452
|
-
}, {
|
|
453
|
-
huggingFace: {
|
|
454
|
-
model: "bartowski/codegemma-2b-GGUF",
|
|
455
|
-
branch: "main",
|
|
456
|
-
file: "codegemma-2b-Q5_K_S.gguf"
|
|
457
|
-
}
|
|
458
|
-
}, {
|
|
459
|
-
huggingFace: {
|
|
460
|
-
model: "bartowski/codegemma-2b-GGUF",
|
|
461
|
-
branch: "main",
|
|
462
|
-
file: "codegemma-2b-Q4_K_M.gguf"
|
|
463
|
-
}
|
|
464
|
-
}]
|
|
185
|
+
fileOptions: [
|
|
186
|
+
"hf:bartowski/codegemma-2b-GGUF/codegemma-2b-Q8_0.gguf",
|
|
187
|
+
"hf:bartowski/codegemma-2b-GGUF/codegemma-2b-Q6_K.gguf",
|
|
188
|
+
"hf:bartowski/codegemma-2b-GGUF/codegemma-2b-Q5_K_M.gguf",
|
|
189
|
+
"hf:bartowski/codegemma-2b-GGUF/codegemma-2b-Q5_K_S.gguf",
|
|
190
|
+
"hf:bartowski/codegemma-2b-GGUF/codegemma-2b-Q4_K_M.gguf"
|
|
191
|
+
]
|
|
465
192
|
}, {
|
|
466
193
|
name: "CodeGemma 7B",
|
|
467
194
|
abilities: ["code", "complete", "infill"],
|
|
@@ -469,47 +196,19 @@ export const recommendedModels = [{
|
|
|
469
196
|
"natual language understanding, mathematical reasoning, and instruction following.\n" +
|
|
470
197
|
"This model is not suited for chat.\n" +
|
|
471
198
|
"This is the 7 billion parameters version of the model.\n",
|
|
472
|
-
fileOptions: [
|
|
473
|
-
|
|
474
|
-
|
|
475
|
-
|
|
476
|
-
|
|
477
|
-
|
|
478
|
-
}, {
|
|
479
|
-
huggingFace: {
|
|
480
|
-
model: "bartowski/codegemma-1.1-7b-it-GGUF",
|
|
481
|
-
branch: "main",
|
|
482
|
-
file: "codegemma-1.1-7b-it-Q5_K_M.gguf"
|
|
483
|
-
}
|
|
484
|
-
}, {
|
|
485
|
-
huggingFace: {
|
|
486
|
-
model: "bartowski/codegemma-1.1-7b-it-GGUF",
|
|
487
|
-
branch: "main",
|
|
488
|
-
file: "codegemma-1.1-7b-it-Q5_K_S.gguf"
|
|
489
|
-
}
|
|
490
|
-
}, {
|
|
491
|
-
huggingFace: {
|
|
492
|
-
model: "bartowski/codegemma-1.1-7b-it-GGUF",
|
|
493
|
-
branch: "main",
|
|
494
|
-
file: "codegemma-1.1-7b-it-Q4_K_M.gguf"
|
|
495
|
-
}
|
|
496
|
-
}]
|
|
199
|
+
fileOptions: [
|
|
200
|
+
"hf:bartowski/codegemma-1.1-7b-it-GGUF/codegemma-1.1-7b-it-Q6_K.gguf",
|
|
201
|
+
"hf:bartowski/codegemma-1.1-7b-it-GGUF/codegemma-1.1-7b-it-Q5_K_M.gguf",
|
|
202
|
+
"hf:bartowski/codegemma-1.1-7b-it-GGUF/codegemma-1.1-7b-it-Q5_K_S.gguf",
|
|
203
|
+
"hf:bartowski/codegemma-1.1-7b-it-GGUF/codegemma-1.1-7b-it-Q4_K_M.gguf"
|
|
204
|
+
]
|
|
497
205
|
}, {
|
|
498
206
|
name: "Stable Code Instruct 3B",
|
|
499
207
|
abilities: ["chat", "complete", "infill"],
|
|
500
208
|
description: "Stable Code models were created by Stability AI and are optimized for code completion.",
|
|
501
|
-
fileOptions: [
|
|
502
|
-
|
|
503
|
-
|
|
504
|
-
|
|
505
|
-
file: "stable-code-3b-q5_k_m.gguf"
|
|
506
|
-
}
|
|
507
|
-
}, {
|
|
508
|
-
huggingFace: {
|
|
509
|
-
model: "stabilityai/stable-code-instruct-3b",
|
|
510
|
-
branch: "main",
|
|
511
|
-
file: "stable-code-3b-q4_k_m.gguf"
|
|
512
|
-
}
|
|
513
|
-
}]
|
|
209
|
+
fileOptions: [
|
|
210
|
+
"hf:stabilityai/stable-code-instruct-3b/stable-code-3b-q5_k_m.gguf",
|
|
211
|
+
"hf:stabilityai/stable-code-instruct-3b/stable-code-3b-q4_k_m.gguf"
|
|
212
|
+
]
|
|
514
213
|
}];
|
|
515
214
|
//# sourceMappingURL=recommendedModels.js.map
|