local_models/models.json at main · Fabrizio1990/local_models · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
{
  "models": [
    {
      "id": "gemma3-1b-it",
      "name": "Gemma 3 1B IT",
      "description": "A variant of google/Gemma-3-1B-IT with 4-bit quantization ready for deployment on Android using the MediaPipe LLM Inference API",
      "size": "557 MB",
      "minRamGB": 2,
      "maxTokens": 4096,
      "maxTokensByRam": {
        "low": 2048,
        "normal": 4096,
        "high": 4096
      },
      "downloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v1.0.0-models/gemma3-1b-it-int4.litertlm",
      "fileName": "gemma3-1b-it-int4.litertlm",
      "format": "litertlm",
      "supportsVision": false,
      "supportsThinking": false
    },
    {
      "id": "gemma-3n-e2b-it",
      "name": "Gemma 3n E2B IT",
      "description": "Preview version of Gemma 3n E2B ready for deployment on Android. Supports text, vision, and audio input with 4096 context length.",
      "size": "3.2 GB",
      "minRamGB": 4,
      "maxTokens": 4096,
      "maxTokensByRam": {
        "low": 2048,
        "normal": 4096,
        "high": 8192
      },
      "downloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v1.0.0-models/gemma-3n-E2B-it-int4.part1.litertlm",
      "fileName": "gemma-3n-E2B-it-int4.litertlm",
      "isMultiPart": true,
      "parts": [
        "gemma-3n-E2B-it-int4.part1.litertlm",
        "gemma-3n-E2B-it-int4.part2.litertlm"
      ],
      "format": "litertlm",
      "supportsVision": false,
      "supportsThinking": false
    },
    {
      "id": "qwen3.5-0.8b-q4km",
      "name": "Qwen 3.5 0.8B",
      "description": "Qwen 3.5 0.8B with Q4_K_M quantization for on-device inference via llama.cpp",
      "size": "533 MB",
      "minRamGB": 2,
      "maxTokens": 4096,
      "maxTokensByRam": {
        "low": 4096,
        "normal": 8192,
        "high": 16384
      },
      "downloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v2.0.0-models/Qwen3.5-0.8B-Q4_K_M.gguf",
      "fileName": "Qwen3.5-0.8B-Q4_K_M.gguf",
      "format": "gguf",
      "supportsVision": true,
      "supportsThinking": true,
      "visionProjectorFileName": "mmproj-qwen3.5-0.8b-F16.gguf",
      "visionProjectorSize": 204987232,
      "visionProjectorDownloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v2.0.0-models/mmproj-qwen3.5-0.8b-F16.gguf"
    },
    {
      "id": "gemma-4-e2b-it-q4km",
      "name": "Gemma 4 E2B",
      "description": "Gemma 4 E2B optimized for MediaPipe (LiteRT) on-device inference. High-performance reasoning on mobile.",
      "size": "3.2 GB",
      "minRamGB": 4,
      "maxTokens": 4096,
      "maxTokensByRam": {
        "low": 2048,
        "normal": 4096,
        "high": 8192
      },
      "downloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v3.1.0-models/gemma-4-E2B-it-int4.part1.litertlm",
      "fileName": "gemma-4-E2B-it-int4.litertlm",
      "format": "litertlm",
      "isMultiPart": true,
      "parts": [
        "gemma-4-E2B-it-int4.part1.litertlm",
        "gemma-4-E2B-it-int4.part2.litertlm"
      ],
      "supportsVision": true,
      "supportsThinking": true,
      "visionProjectorFileName": "mmproj-google_gemma-4-E2B-it-f16.gguf",
      "visionProjectorSize": 985653760,
      "visionProjectorDownloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v3.0.0-models/mmproj-google_gemma-4-E2B-it-f16.gguf"
    },
    {
      "id": "gemma-4-e4b-it-q4km",
      "name": "Gemma 4 E4B",
      "description": "Gemma 4 E4B optimized for MediaPipe (LiteRT) on-device inference. Superior reasoning for flagship devices.",
      "size": "5.4 GB",
      "minRamGB": 8,
      "maxTokens": 4096,
      "maxTokensByRam": {
        "low": 2048,
        "normal": 4096,
        "high": 6144
      },
      "downloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v3.1.0-models/gemma-4-E4B-it-int4.part1.litertlm",
      "fileName": "gemma-4-E4B-it-int4.litertlm",
      "format": "litertlm",
      "isMultiPart": true,
      "parts": [
        "gemma-4-E4B-it-int4.part1.litertlm",
        "gemma-4-E4B-it-int4.part2.litertlm"
      ],
      "supportsVision": true,
      "supportsThinking": true,
      "visionProjectorFileName": "mmproj-google_gemma-4-E4B-it-f16.gguf",
      "visionProjectorSize": 990372352,
      "visionProjectorDownloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v3.0.0-models/mmproj-google_gemma-4-E4B-it-f16.gguf"
    },
    {
      "id": "qwen3.5-2b-q4km",
      "name": "Qwen 3.5 2B",
      "description": "Qwen 3.5 2B with Q4_K_M quantization for on-device inference via llama.cpp",
      "size": "1.28 GB",
      "minRamGB": 4,
      "maxTokens": 4096,
      "maxTokensByRam": {
        "low": 2048,
        "normal": 4096,
        "high": 8192
      },
      "downloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v2.0.0-models/Qwen3.5-2B-Q4_K_M.gguf",
      "fileName": "Qwen3.5-2B-Q4_K_M.gguf",
      "format": "gguf",
      "supportsVision": true,
      "supportsThinking": true,
      "visionProjectorFileName": "mmproj-qwen3.5-2b-F16.gguf",
      "visionProjectorSize": 668227264,
      "visionProjectorDownloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v2.0.0-models/mmproj-qwen3.5-2b-F16.gguf"
    },
    {
      "id": "qwen3.5-4b-q4km",
      "name": "Qwen 3.5 4B",
      "description": "Qwen 3.5 4B with Q4_K_M quantization for on-device inference via llama.cpp",
      "size": "2.74 GB",
      "minRamGB": 6,
      "maxTokens": 4096,
      "maxTokensByRam": {
        "low": 2048,
        "normal": 4096,
        "high": 6144
      },
      "downloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v2.0.0-models/Qwen3.5-4B-Q4_K_M.part1.gguf",
      "fileName": "Qwen3.5-4B-Q4_K_M.gguf",
      "isMultiPart": true,
      "parts": [
        "Qwen3.5-4B-Q4_K_M.part1.gguf",
        "Qwen3.5-4B-Q4_K_M.part2.gguf"
      ],
      "format": "gguf",
      "supportsVision": true,
      "supportsThinking": true,
      "visionProjectorFileName": "mmproj-qwen3.5-4b-F16.gguf",
      "visionProjectorSize": 672423616,
      "visionProjectorDownloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v2.0.0-models/mmproj-qwen3.5-4b-F16.gguf"
    },
    {
      "id": "qwen3.5-9b-q4km",
      "name": "Qwen 3.5 9B",
      "description": "Qwen 3.5 9B with Q4_K_M quantization for on-device inference via llama.cpp",
      "size": "5.68 GB",
      "minRamGB": 10,
      "maxTokens": 4096,
      "maxTokensByRam": {
        "low": 2048,
        "normal": 2048,
        "high": 4096
      },
      "downloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v2.0.0-models/Qwen3.5-9B-Q4_K_M.part1.gguf",
      "fileName": "Qwen3.5-9B-Q4_K_M.gguf",
      "isMultiPart": true,
      "parts": [
        "Qwen3.5-9B-Q4_K_M.part1.gguf",
        "Qwen3.5-9B-Q4_K_M.part2.gguf",
        "Qwen3.5-9B-Q4_K_M.part3.gguf"
      ],
      "format": "gguf",
      "supportsVision": true,
      "supportsThinking": true,
      "visionProjectorFileName": "mmproj-qwen3.5-9b-F16.gguf",
      "visionProjectorSize": 918166080,
      "visionProjectorDownloadUrl": "https://github.com/Fabrizio1990/local_models/releases/download/v2.0.0-models/mmproj-qwen3.5-9b-F16.gguf"
    }
  ]
}