Skip to content

Commit b64347b

Browse files
mudlermudler
committed
chore: add gemma4 to the gallery
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
1 parent e00ce98 commit b64347b

1 file changed

Lines changed: 128 additions & 0 deletions

File tree

‎gallery/index.yaml‎

Lines changed: 128 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -27,6 +27,134 @@
2727
- filename: llama-cpp/models/gemma-4-26B-A4B-it-APEX-GGUF/gemma-4-26B-A4B-APEX-Quality.gguf
2828
uri: https://huggingface.co/mudler/gemma-4-26B-A4B-it-APEX-GGUF/resolve/main/gemma-4-26B-A4B-APEX-Quality.gguf
2929
sha256: a6591d7b41978e6f465acd9d03e96286f70912402c695158fb267ccbfbb740ed
30+
- &gemma4
31+
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
32+
name: "gemma-4-26b-a4b-it"
33+
icon: https://ai.google.dev/static/gemma/images/gemma3.png
34+
license: gemma
35+
urls:
36+
- https://huggingface.co/google/gemma-4-26B-A4B-it
37+
- https://huggingface.co/ggml-org/gemma-4-26B-A4B-it-GGUF
38+
description: |
39+
Google Gemma 4 26B-A4B-IT is an open-source multimodal Mixture-of-Experts model with 26B total parameters and 4B active parameters. It handles text and image input, generating text output, with a 256K context window and support for 140+ languages. The MoE architecture provides strong performance with efficient inference. Well-suited for question answering, summarization, reasoning, and image understanding tasks.
40+
tags:
41+
- llm
42+
- gguf
43+
- gpu
44+
- cpu
45+
- gemma
46+
- gemma4
47+
- gemma-4
48+
- multimodal
49+
overrides:
50+
backend: llama-cpp
51+
function:
52+
automatic_tool_parsing_fallback: true
53+
grammar:
54+
disable: true
55+
known_usecases:
56+
- chat
57+
mmproj: mmproj-gemma-4-26B-A4B-it-f16.gguf
58+
options:
59+
- use_jinja:true
60+
parameters:
61+
model: gemma-4-26B-A4B-it-Q4_K_M.gguf
62+
template:
63+
use_tokenizer_template: true
64+
files:
65+
- filename: gemma-4-26B-A4B-it-Q4_K_M.gguf
66+
sha256: 23c6997912cb7fa36147fe05877de73ddbb2a80ff69b18ff171b354dccf2b5b5
67+
uri: huggingface://ggml-org/gemma-4-26B-A4B-it-GGUF/gemma-4-26B-A4B-it-Q4_K_M.gguf
68+
- filename: mmproj-gemma-4-26B-A4B-it-f16.gguf
69+
sha256: 4107c1c3c299095fbc323f87f4e4cac81dd9527db5ff90808fea669e08244531
70+
uri: huggingface://ggml-org/gemma-4-26B-A4B-it-GGUF/mmproj-gemma-4-26B-A4B-it-f16.gguf
71+
- !!merge <<: *gemma4
72+
name: "gemma-4-e2b-it"
73+
urls:
74+
- https://huggingface.co/google/gemma-4-E2B-it
75+
- https://huggingface.co/ggml-org/gemma-4-E2B-it-GGUF
76+
description: |
77+
Google Gemma 4 E2B-IT is a lightweight open-source multimodal model with 5B total parameters and 2B effective parameters using selective parameter activation. It handles text and image input, generating text output, with a 256K context window and support for 140+ languages. Optimized for efficient execution on low-resource devices including mobile and laptops.
78+
overrides:
79+
backend: llama-cpp
80+
function:
81+
automatic_tool_parsing_fallback: true
82+
grammar:
83+
disable: true
84+
known_usecases:
85+
- chat
86+
mmproj: mmproj-gemma-4-e2b-it-f16.gguf
87+
options:
88+
- use_jinja:true
89+
parameters:
90+
model: gemma-4-e2b-it-Q8_0.gguf
91+
template:
92+
use_tokenizer_template: true
93+
files:
94+
- filename: gemma-4-e2b-it-Q8_0.gguf
95+
sha256: 12d878964d21f1779dea15abeee048855151b27089fe98b32c628f85740933f3
96+
uri: huggingface://ggml-org/gemma-4-E2B-it-GGUF/gemma-4-e2b-it-Q8_0.gguf
97+
- filename: mmproj-gemma-4-e2b-it-f16.gguf
98+
sha256: 9165f3d9674c3731ae29373d95b860d141eee030b0ec0bf4577e2de8596a7767
99+
uri: huggingface://ggml-org/gemma-4-E2B-it-GGUF/mmproj-gemma-4-e2b-it-f16.gguf
100+
- !!merge <<: *gemma4
101+
name: "gemma-4-e4b-it"
102+
urls:
103+
- https://huggingface.co/google/gemma-4-E4B-it
104+
- https://huggingface.co/ggml-org/gemma-4-E4B-it-GGUF
105+
description: |
106+
Google Gemma 4 E4B-IT is an open-source multimodal model with 8B total parameters and 4B effective parameters using selective parameter activation. It handles text and image input, generating text output, with a 256K context window and support for 140+ languages. Offers a good balance of performance and efficiency for deployment on consumer hardware.
107+
overrides:
108+
backend: llama-cpp
109+
function:
110+
automatic_tool_parsing_fallback: true
111+
grammar:
112+
disable: true
113+
known_usecases:
114+
- chat
115+
mmproj: mmproj-gemma-4-e4b-it-f16.gguf
116+
options:
117+
- use_jinja:true
118+
parameters:
119+
model: gemma-4-e4b-it-Q4_K_M.gguf
120+
template:
121+
use_tokenizer_template: true
122+
files:
123+
- filename: gemma-4-e4b-it-Q4_K_M.gguf
124+
sha256: dff4e4ca848e33e678a63b5b7d1f8bfa4a17e764415d0c0aaaad07c84f4d8fad
125+
uri: huggingface://ggml-org/gemma-4-E4B-it-GGUF/gemma-4-e4b-it-Q4_K_M.gguf
126+
- filename: mmproj-gemma-4-e4b-it-f16.gguf
127+
sha256: a7e94f39cee4569fae49c852cbfb574c54e225aafb8b75313a8bf06b89e17712
128+
uri: huggingface://ggml-org/gemma-4-E4B-it-GGUF/mmproj-gemma-4-e4b-it-f16.gguf
129+
- !!merge <<: *gemma4
130+
name: "gemma-4-31b-it"
131+
urls:
132+
- https://huggingface.co/google/gemma-4-31B-it
133+
- https://huggingface.co/unsloth/gemma-4-31B-it-GGUF
134+
description: |
135+
Google Gemma 4 31B-IT is the largest dense model in the Gemma 4 family with 31B parameters. It handles text and image input, generating text output, with a 256K context window and support for 140+ languages. Provides the highest quality outputs in the Gemma 4 lineup, well-suited for complex reasoning, summarization, and image understanding tasks.
136+
overrides:
137+
backend: llama-cpp
138+
function:
139+
automatic_tool_parsing_fallback: true
140+
grammar:
141+
disable: true
142+
known_usecases:
143+
- chat
144+
mmproj: mmproj-F16.gguf
145+
options:
146+
- use_jinja:true
147+
parameters:
148+
model: gemma-4-31B-it-Q4_K_M.gguf
149+
template:
150+
use_tokenizer_template: true
151+
files:
152+
- filename: gemma-4-31B-it-Q4_K_M.gguf
153+
sha256: 5783acab7217360984ea957abc36f89d35cdeba2f5dd30b0e9e33f9f294bad82
154+
uri: huggingface://unsloth/gemma-4-31B-it-GGUF/gemma-4-31B-it-Q4_K_M.gguf
155+
- filename: mmproj-F16.gguf
156+
sha256: 1be2a32013e0d29c6c746513b5f5e7d38f47b694351e74a69c7172acdbbb11a6
157+
uri: https://huggingface.co/unsloth/gemma-4-31B-it-GGUF/resolve/main/mmproj-F16.gguf
30158
- name: "qwen3.5-35b-a3b-apex"
31159
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
32160
urls:

0 commit comments

Comments
 (0)