mudler · github-actions · Mar 30, 2025 · Mar 30, 2025
diff --git a/Makefile b/Makefile
@@ -6,7 +6,7 @@ BINARY_NAME=local-ai
 DETECT_LIBS?=true
 
 # llama.cpp versions
-CPPLLAMA_VERSION?=0bb2919335d00ff0bc79d5015da95c422de51f03
+CPPLLAMA_VERSION?=4663bd353c61c1136cd8a97b9908755e4ab30cec
 
 # whisper.cpp version
 WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp

diff --git a/gallery/index.yaml b/gallery/index.yaml
    - gemma3
    - gemma-3
  overrides:
    #mmproj: gemma-3-27b-it-mmproj-f16.gguf
    parameters:
      model: gemma-3-27b-it-Q4_K_M.gguf
  files:
  description: |
    google/gemma-3-12b-it is an open-source, state-of-the-art, lightweight, multimodal model built from the same research and technology used to create the Gemini models. It is capable of handling text and image input and generating text output. It has a large context window of 128K tokens and supports over 140 languages. The 12B variant has been fine-tuned using the instruction-tuning approach. Gemma 3 models are suitable for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes them deployable in environments with limited resources such as laptops, desktops, or your own cloud infrastructure.
  overrides:
    #mmproj: gemma-3-12b-it-mmproj-f16.gguf
    parameters:
      model: gemma-3-12b-it-Q4_K_M.gguf
  files:
  description: |
    Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. Gemma 3 models are multimodal, handling text and image input and generating text output, with open weights for both pre-trained variants and instruction-tuned variants. Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions. Gemma 3 models are well-suited for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes it possible to deploy them in environments with limited resources such as laptops, desktops or your own cloud infrastructure, democratizing access to state of the art AI models and helping foster innovation for everyone. Gemma-3-4b-it is a 4 billion parameter model.
  overrides:
    #mmproj: gemma-3-4b-it-mmproj-f16.gguf
    parameters:
      model: gemma-3-4b-it-Q4_K_M.gguf
  files:
      sha256: 8a8e7a0fa1068755322c51900e53423d795e57976b4d95982242cbec41141c7b
      uri: huggingface://mradermacher/Eximius_Persona_5B-GGUF/Eximius_Persona_5B.Q4_K_M.gguf
 - &qwen25
  name: "qwen2.5-14b-instruct" ## Qwen2.5
  icon: https://avatars.githubusercontent.com/u/141221163
  url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
  license: apache-2.0
      sha256: 0fec82625f74a9a340837de7af287b1d9042e5aeb70cda2621426db99958b0af
      uri: huggingface://bartowski/Chuluun-Qwen2.5-72B-v0.08-GGUF/Chuluun-Qwen2.5-72B-v0.08-Q4_K_M.gguf
 - &smollm
  url: "github:mudler/LocalAI/gallery/chatml.yaml@master" ## SmolLM
  name: "smollm-1.7b-instruct"
  icon: https://huggingface.co/datasets/HuggingFaceTB/images/resolve/main/banner_smol.png
  tags:
      sha256: 3682843c857595765f0786cf24b3d501af96fe5d99a9fb2526bc7707e28bae1e
      uri: huggingface://Nekuromento/Hammer2.0-7b-Q5_K_M-GGUF/hammer2.0-7b-q5_k_m.gguf
 - &llama31
  url: "github:mudler/LocalAI/gallery/llama3.1-instruct.yaml@master" ## LLama3.1
  icon: https://avatars.githubusercontent.com/u/153379578
  name: "meta-llama-3.1-8b-instruct"
  license: llama3.1
      sha256: 25db6d4ae5649e6d2084036d8f05ec1aca459126e2d4734d6c18f1e16147a4d3
      uri: huggingface://mradermacher/Llama-3.3-MagicalGirl-2.5-i1-GGUF/Llama-3.3-MagicalGirl-2.5.i1-Q4_K_M.gguf
 - &deepseek
  url: "github:mudler/LocalAI/gallery/deepseek.yaml@master" ## Deepseek
  name: "deepseek-coder-v2-lite-instruct"
  icon: "https://avatars.githubusercontent.com/u/148330874"
  license: deepseek
      sha256: a47782c55ef2b39b19644213720a599d9849511a73c9ebb0c1de749383c0a0f8
      uri: huggingface://RichardErkhov/ContextualAI_-_archangel_sft_pythia2-8b-gguf/archangel_sft_pythia2-8b.Q4_K_M.gguf
 - &deepseek-r1
  url: "github:mudler/LocalAI/gallery/deepseek-r1.yaml@master" ## Start DeepSeek-R1
  name: "deepseek-r1-distill-qwen-1.5b"
  icon: "https://avatars.githubusercontent.com/u/148330874"
  urls:
      sha256: aed6bd5bb03b7bd886939237bc10ea6331d4feb5a3b6712e0c5474a778acf817
      uri: huggingface://mradermacher/Fallen-Safeword-70B-R1-v4.1-GGUF/Fallen-Safeword-70B-R1-v4.1.Q4_K_M.gguf
 - &qwen2
  url: "github:mudler/LocalAI/gallery/chatml.yaml@master" ## Start QWEN2
  name: "qwen2-7b-instruct"
  icon: https://avatars.githubusercontent.com/u/141221163
  license: apache-2.0
	- gemma3
	- gemma-3
	overrides:
	#mmproj: gemma-3-27b-it-mmproj-f16.gguf
Check warning on line 21 in gallery/index.yaml GitHub Actions / Yamllint `21:6 [comments] missing starting space in comment`
	parameters:
	model: gemma-3-27b-it-Q4_K_M.gguf
	files:
	description: \|
	google/gemma-3-12b-it is an open-source, state-of-the-art, lightweight, multimodal model built from the same research and technology used to create the Gemini models. It is capable of handling text and image input and generating text output. It has a large context window of 128K tokens and supports over 140 languages. The 12B variant has been fine-tuned using the instruction-tuning approach. Gemma 3 models are suitable for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes them deployable in environments with limited resources such as laptops, desktops, or your own cloud infrastructure.
	overrides:
	#mmproj: gemma-3-12b-it-mmproj-f16.gguf
Check warning on line 39 in gallery/index.yaml GitHub Actions / Yamllint `39:6 [comments] missing starting space in comment`
	parameters:
	model: gemma-3-12b-it-Q4_K_M.gguf
	files:
	description: \|
	Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. Gemma 3 models are multimodal, handling text and image input and generating text output, with open weights for both pre-trained variants and instruction-tuned variants. Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions. Gemma 3 models are well-suited for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes it possible to deploy them in environments with limited resources such as laptops, desktops or your own cloud infrastructure, democratizing access to state of the art AI models and helping foster innovation for everyone. Gemma-3-4b-it is a 4 billion parameter model.
	overrides:
	#mmproj: gemma-3-4b-it-mmproj-f16.gguf
Check warning on line 57 in gallery/index.yaml GitHub Actions / Yamllint `57:6 [comments] missing starting space in comment`
	parameters:
	model: gemma-3-4b-it-Q4_K_M.gguf
	files:
	sha256: 8a8e7a0fa1068755322c51900e53423d795e57976b4d95982242cbec41141c7b
	uri: huggingface://mradermacher/Eximius_Persona_5B-GGUF/Eximius_Persona_5B.Q4_K_M.gguf
	- &qwen25
	name: "qwen2.5-14b-instruct" ## Qwen2.5
Check warning on line 2499 in gallery/index.yaml GitHub Actions / Yamllint `2499:32 [comments] too few spaces before comment`
	icon: https://avatars.githubusercontent.com/u/141221163
	url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
	license: apache-2.0
	sha256: 0fec82625f74a9a340837de7af287b1d9042e5aeb70cda2621426db99958b0af
	uri: huggingface://bartowski/Chuluun-Qwen2.5-72B-v0.08-GGUF/Chuluun-Qwen2.5-72B-v0.08-Q4_K_M.gguf
	- &smollm
	url: "github:mudler/LocalAI/gallery/chatml.yaml@master" ## SmolLM
Check warning on line 4460 in gallery/index.yaml GitHub Actions / Yamllint `4460:59 [comments] too few spaces before comment`
	name: "smollm-1.7b-instruct"
	icon: https://huggingface.co/datasets/HuggingFaceTB/images/resolve/main/banner_smol.png
	tags:
	sha256: 3682843c857595765f0786cf24b3d501af96fe5d99a9fb2526bc7707e28bae1e
	uri: huggingface://Nekuromento/Hammer2.0-7b-Q5_K_M-GGUF/hammer2.0-7b-q5_k_m.gguf
	- &llama31
	url: "github:mudler/LocalAI/gallery/llama3.1-instruct.yaml@master" ## LLama3.1
Check warning on line 5414 in gallery/index.yaml GitHub Actions / Yamllint `5414:70 [comments] too few spaces before comment`
	icon: https://avatars.githubusercontent.com/u/153379578
	name: "meta-llama-3.1-8b-instruct"
	license: llama3.1
	sha256: 25db6d4ae5649e6d2084036d8f05ec1aca459126e2d4734d6c18f1e16147a4d3
	uri: huggingface://mradermacher/Llama-3.3-MagicalGirl-2.5-i1-GGUF/Llama-3.3-MagicalGirl-2.5.i1-Q4_K_M.gguf
	- &deepseek
	url: "github:mudler/LocalAI/gallery/deepseek.yaml@master" ## Deepseek
Check warning on line 7575 in gallery/index.yaml GitHub Actions / Yamllint `7575:61 [comments] too few spaces before comment`
	name: "deepseek-coder-v2-lite-instruct"
	icon: "https://avatars.githubusercontent.com/u/148330874"
	license: deepseek
	sha256: a47782c55ef2b39b19644213720a599d9849511a73c9ebb0c1de749383c0a0f8
	uri: huggingface://RichardErkhov/ContextualAI_-_archangel_sft_pythia2-8b-gguf/archangel_sft_pythia2-8b.Q4_K_M.gguf
	- &deepseek-r1
	url: "github:mudler/LocalAI/gallery/deepseek-r1.yaml@master" ## Start DeepSeek-R1
Check warning on line 7640 in gallery/index.yaml GitHub Actions / Yamllint `7640:64 [comments] too few spaces before comment`
	name: "deepseek-r1-distill-qwen-1.5b"
	icon: "https://avatars.githubusercontent.com/u/148330874"
	urls:
	sha256: aed6bd5bb03b7bd886939237bc10ea6331d4feb5a3b6712e0c5474a778acf817
	uri: huggingface://mradermacher/Fallen-Safeword-70B-R1-v4.1-GGUF/Fallen-Safeword-70B-R1-v4.1.Q4_K_M.gguf
	- &qwen2
	url: "github:mudler/LocalAI/gallery/chatml.yaml@master" ## Start QWEN2
Check warning on line 8030 in gallery/index.yaml GitHub Actions / Yamllint `8030:59 [comments] too few spaces before comment`
	name: "qwen2-7b-instruct"
	icon: https://avatars.githubusercontent.com/u/141221163
	license: apache-2.0