feat(gallery): add Gemma 4 Scotoma 2 variants (#11458)

Add Q4_K_M and Q8_0 multimodal builds with the matching vision projector.

Assisted-by: Codex:gpt-5 [HF API]

Co-authored-by: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com>
This commit is contained in:
localai-org-maint-botandlocalai-org-maint-bot authored and GitHub committed 2026-08-11 09:47:40 +02:00
1 parent 8ad22eeaef
commit fbe1d5036d
1 file changed
+74
+74
View File
@@ -4407,6 +4407,80 @@
- filename: llama-cpp/mmproj/Qwopus3.6-27B-Coder-MTP-GGUF/mmproj-F32.gguf
sha256: 32f7ea0600c07272547da401d460f8abbd980f3a57b69d6df87be0e2505e0b9c
uri: https://huggingface.co/Jackrong/Qwopus3.6-27B-Coder-MTP-GGUF/resolve/main/mmproj-F32.gguf
- &gemma-4-31b-scotoma-2
name: "gemma-4-31b-scotoma-2-q4"
variants:
- model: gemma-4-31b-scotoma-2-q8
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
urls:
- https://huggingface.co/ReadyArt/gemma-4-31B-it-scotoma-2
- https://huggingface.co/ReadyArt/gemma-4-31B-it-scotoma-2-GGUF
description: |
Gemma 4 31B Scotoma 2 is a multimodal Gemma 4 31B instruction-tuned model from ReadyArt. It applies a bounded refusal edit and preference training intended to reduce repetitive prose patterns while retaining the base model's text and image capabilities. This entry uses the 18.7 GB Q4_K_M GGUF and the matching Q8_0 vision projector.
License: Apache 2.0 | Base model: Google Gemma 4 31B IT
license: "apache-2.0"
icon: https://ai.google.dev/gemma/images/gemma4_banner.png
tags:
- llm
- gguf
- gemma
- gemma4
- multimodal
- reasoning
overrides:
backend: llama-cpp
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: llama-cpp/mmproj/gemma-4-31B-it-scotoma-2-GGUF/mmproj-gemma-4-31B-scotoma-2-Q8_0.gguf
options:
- use_jinja:true
parameters:
model: llama-cpp/models/gemma-4-31B-it-scotoma-2-GGUF/gemma-4-31B-scotoma-2-Q4_K_M.gguf
template:
use_tokenizer_template: true
files:
- filename: llama-cpp/models/gemma-4-31B-it-scotoma-2-GGUF/gemma-4-31B-scotoma-2-Q4_K_M.gguf
sha256: 3d2d373d499662cfc5703adcd46b527054267fd750d1e55b11c3d2861b5e7330
uri: huggingface://ReadyArt/gemma-4-31B-it-scotoma-2-GGUF/gemma-4-31B-scotoma-2-Q4_K_M.gguf
- filename: llama-cpp/mmproj/gemma-4-31B-it-scotoma-2-GGUF/mmproj-gemma-4-31B-scotoma-2-Q8_0.gguf
sha256: a65401a0143d4a9883a4bca7e0c780b7140fc98142aaa26790da1ee46aa163dd
uri: huggingface://ReadyArt/gemma-4-31B-it-scotoma-2-GGUF/mmproj-gemma-4-31B-scotoma-2-Q8_0.gguf
- !!merge <<: *gemma-4-31b-scotoma-2
name: "gemma-4-31b-scotoma-2-q8"
variants: []
description: |
Gemma 4 31B Scotoma 2 is a multimodal Gemma 4 31B instruction-tuned model from ReadyArt. It applies a bounded refusal edit and preference training intended to reduce repetitive prose patterns while retaining the base model's text and image capabilities. This higher-fidelity entry uses the 32.6 GB Q8_0 GGUF and the matching Q8_0 vision projector.
License: Apache 2.0 | Base model: Google Gemma 4 31B IT
overrides:
backend: llama-cpp
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: llama-cpp/mmproj/gemma-4-31B-it-scotoma-2-GGUF/mmproj-gemma-4-31B-scotoma-2-Q8_0.gguf
options:
- use_jinja:true
parameters:
model: llama-cpp/models/gemma-4-31B-it-scotoma-2-GGUF/gemma-4-31B-scotoma-2-Q8_0.gguf
template:
use_tokenizer_template: true
files:
- filename: llama-cpp/models/gemma-4-31B-it-scotoma-2-GGUF/gemma-4-31B-scotoma-2-Q8_0.gguf
sha256: 79d9125f9769d95ceaa5e77cca72e9af99f4c9effb5bbe78999e9f3afe5e8d45
uri: huggingface://ReadyArt/gemma-4-31B-it-scotoma-2-GGUF/gemma-4-31B-scotoma-2-Q8_0.gguf
- filename: llama-cpp/mmproj/gemma-4-31B-it-scotoma-2-GGUF/mmproj-gemma-4-31B-scotoma-2-Q8_0.gguf
sha256: a65401a0143d4a9883a4bca7e0c780b7140fc98142aaa26790da1ee46aa163dd
uri: huggingface://ReadyArt/gemma-4-31B-it-scotoma-2-GGUF/mmproj-gemma-4-31B-scotoma-2-Q8_0.gguf
- name: "gemma-4-26b-a4b-it-qat"
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
urls: