fix(api)!: Stop model prior to deletion

.github/workflows/backend.yml

@@ -14,6 +14,7 @@ concurrency:
 
 jobs:
   backend-jobs:
+    if: github.repository == 'mudler/LocalAI'
     uses: ./.github/workflows/backend_build.yml
     with:
       tag-latest: ${{ matrix.tag-latest }}

.github/workflows/bump_deps.yaml

@@ -5,6 +5,7 @@ on:
   workflow_dispatch:
 jobs:
   bump-backends:
+    if: github.repository == 'mudler/LocalAI'
     strategy:
       fail-fast: false
       matrix:

.github/workflows/bump_docs.yaml

@@ -5,6 +5,7 @@ on:
   workflow_dispatch:
 jobs:
   bump-docs:
+    if: github.repository == 'mudler/LocalAI'
     strategy:
       fail-fast: false
       matrix:

.github/workflows/checksum_checker.yaml

@@ -5,6 +5,7 @@ on:
   workflow_dispatch:
 jobs:
   checksum_check:
+    if: github.repository == 'mudler/LocalAI'
     runs-on: ubuntu-latest
     steps:
       - name: Force Install GIT latest

.github/workflows/dependabot_auto.yml

@@ -9,8 +9,8 @@ permissions:
 
 jobs:
   dependabot:
+    if: github.repository == 'mudler/LocalAI' && github.actor == 'dependabot[bot]'
     runs-on: ubuntu-latest
-    if: ${{ github.actor == 'dependabot[bot]' }}
     steps:
       - name: Dependabot metadata
         id: metadata

.github/workflows/deploy-explorer.yaml

@@ -12,6 +12,7 @@ concurrency:
 
 jobs:
   build-linux:
+    if: github.repository == 'mudler/LocalAI'
     runs-on: ubuntu-latest
     steps:
       - name: Clone

.github/workflows/gallery-agent.yaml

@@ -27,6 +27,7 @@ on:
         type: string
 jobs:
   gallery-agent:
+    if: github.repository == 'mudler/LocalAI'
     runs-on: ubuntu-latest
     steps:
       - name: Checkout repository

.github/workflows/generate_grpc_cache.yaml

@@ -13,6 +13,7 @@ concurrency:
 
 jobs:
   generate_caches:
+    if: github.repository == 'mudler/LocalAI'
     strategy:
       matrix:
         include:

.github/workflows/generate_intel_image.yaml

@@ -12,6 +12,7 @@ concurrency:
 
 jobs:
   generate_caches:
+    if: github.repository == 'mudler/LocalAI'
     strategy:
       matrix:
         include:

.github/workflows/image.yml

@@ -14,6 +14,7 @@
 
   jobs:
     hipblas-jobs:
+      if: github.repository == 'mudler/LocalAI'
       uses: ./.github/workflows/image_build.yml
       with:
         tag-latest: ${{ matrix.tag-latest }}
@@ -50,6 +51,7 @@
               ubuntu-codename: 'noble'
 
     core-image-build:
+      if: github.repository == 'mudler/LocalAI'
       uses: ./.github/workflows/image_build.yml
       with:
         tag-latest: ${{ matrix.tag-latest }}
@@ -136,6 +138,7 @@
               ubuntu-codename: 'noble'
 
     gh-runner:
+      if: github.repository == 'mudler/LocalAI'
       uses: ./.github/workflows/image_build.yml
       with:
         tag-latest: ${{ matrix.tag-latest }}

.github/workflows/localaibot_automerge.yml

@@ -10,8 +10,8 @@ permissions:
   actions: write # to dispatch publish workflow
 jobs:
   dependabot:
+    if: github.repository == 'mudler/LocalAI' && github.actor == 'localai-bot' && !contains(github.event.pull_request.title, 'chore(model gallery):')
     runs-on: ubuntu-latest
-    if: ${{ github.actor == 'localai-bot' && !contains(github.event.pull_request.title, 'chore(model gallery):') }}
     steps:
       - name: Checkout repository
         uses: actions/checkout@v6

.github/workflows/notify-models.yaml

@@ -10,7 +10,7 @@ permissions:
 
 jobs:
   notify-discord:
-    if: ${{ (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model')) }}
+    if: github.repository == 'mudler/LocalAI' && (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model'))
     env:
         MODEL_NAME: gemma-3-12b-it-qat
     runs-on: ubuntu-latest
@@ -90,7 +90,7 @@ jobs:
         connect-timeout-seconds: 180
         limit-access-to-actor: true
   notify-twitter:
-    if: ${{ (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model')) }}
+    if: github.repository == 'mudler/LocalAI' && (github.event.pull_request.merged == true) && (contains(github.event.pull_request.labels.*.name, 'area/ai-model'))
     env:
         MODEL_NAME: gemma-3-12b-it-qat
     runs-on: ubuntu-latest

.github/workflows/notify-releases.yaml

@@ -6,6 +6,7 @@ on:
 
 jobs:
   notify-discord:
+    if: github.repository == 'mudler/LocalAI'
     runs-on: ubuntu-latest
     env:
         RELEASE_BODY: ${{ github.event.release.body }}

.github/workflows/stalebot.yml

@@ -8,6 +8,7 @@ on:
 
 jobs:
   stale:
+    if: github.repository == 'mudler/LocalAI'
     runs-on: ubuntu-latest
     steps:
       - uses: actions/stale@997185467fa4f803885201cee163a9f38240193d # v9

.github/workflows/update_swagger.yaml

@@ -5,6 +5,7 @@ on:
   workflow_dispatch:
 jobs:
   swagger:
+    if: github.repository == 'mudler/LocalAI'
     strategy:
       fail-fast: false
     runs-on: ubuntu-latest

backend/go/whisper/Makefile

@@ -78,7 +78,7 @@ package: whisper
 build: package
 
 clean: purge
-	rm -rf libgowhisper*.so sources/whisper.cpp whisper
+	rm -rf libgowhisper*.so package sources/whisper.cpp whisper
 
 purge:
 	rm -rf build*

core/services/models.go

@@ -197,6 +197,9 @@ func processModelOperation(
 
 	switch {
 	case op.Delete:
+		if err := modelLoader.ShutdownModel(op.GalleryElementName); err != nil {
+			xlog.Warn("Failed to unload model during deletion", "model", op.GalleryElementName, "error", err)
+		}
 		return gallery.DeleteModelFromSystem(systemState, op.GalleryElementName)
 	case op.GalleryElement != nil:
 		installedModel, err := gallery.InstallModel(

gallery/index.yaml

@@ -31,7 +31,7 @@
       - "offload_to_cpu:false"
       - "offload_dit_to_cpu:false"
       - "init_lm:true"
       - "lm_model_path:acestep-5Hz-lm-0.6B" # or acestep-5Hz-lm-4B
       - "lm_backend:pt"
       - "temperature:0.85"
       - "top_p:0.9"
@@ -264,7 +264,7 @@
     known_usecases:
       - tts
     tts:
       voice: Aiden # Available speakers: Vivian, Serena, Uncle_Fu, Dylan, Eric, Ryan, Aiden, Ono_Anna, Sohee
     parameters:
       model: Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice
 - &qwen-asr
@@ -1391,7 +1391,7 @@
     - filename: mmproj-OpenGVLab_InternVL3_5-2B-f16.gguf
       sha256: e83ba6e675b747f7801557dc24594f43c17a7850b6129d4972d55e3e9b010359
       uri: huggingface://bartowski/OpenGVLab_InternVL3_5-8B-GGUF/mmproj-OpenGVLab_InternVL3_5-2B-f16.gguf
-- &lfm2
+- &lfm2vl
   url: "github:mudler/LocalAI/gallery/lfm.yaml@master"
   name: "lfm2-vl-450m"
   license: lfm1.0
@@ -1424,7 +1424,7 @@
     - filename: mmproj-LFM2-VL-450M-F16.gguf
       sha256: 416a085c5c7ba0f8d02bb8326c719a6f8f2210c2641c6bf64194a57c11c76e59
       uri: huggingface://LiquidAI/LFM2-VL-450M-GGUF/mmproj-LFM2-VL-450M-F16.gguf
-- !!merge <<: *lfm2
+- !!merge <<: *lfm2vl
   name: "lfm2-vl-1.6b"
   urls:
     - https://huggingface.co/LiquidAI/LFM2-VL-1.6B
@@ -1440,7 +1440,7 @@
     - filename: mmproj-LFM2-VL-1.6B-F16.gguf
       sha256: b637bfa6060be2bc7503ec23ba48b407843d08c2ca83f52be206ea8563ccbae2
       uri: huggingface://LiquidAI/LFM2-VL-1.6B-GGUF/mmproj-LFM2-VL-1.6B-F16.gguf
-- !!merge <<: *lfm2
+- &lfm2
   name: "lfm2-1.2b"
   urls:
     - https://huggingface.co/LiquidAI/LFM2-1.2B
@@ -4305,7 +4305,7 @@
     - gemma3
     - gemma-3
   overrides:
     #mmproj: gemma-3-27b-it-mmproj-f16.gguf
     parameters:
       model: gemma-3-27b-it-Q4_K_M.gguf
   files:
@@ -4323,7 +4323,7 @@
   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:
@@ -4341,7 +4341,7 @@
   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:
@@ -7559,7 +7559,7 @@
       sha256: 2756551de7d8ff7093c2c5eec1cd00f1868bc128433af53f5a8d434091d4eb5a
       uri: huggingface://Triangle104/Nano_Imp_1B-Q8_0-GGUF/nano_imp_1b-q8_0.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:
@@ -7603,7 +7603,7 @@
       sha256: decd2598bc2c8ed08c19adc3c8fdd461ee19ed5708679d1c54ef54a5a30d4f33
       uri: huggingface://HuggingFaceTB/SmolLM2-1.7B-Instruct-GGUF/smollm2-1.7b-instruct-q4_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

pkg/system/capabilities.go

@@ -45,8 +45,9 @@ const (
 )
 
 var (
-	cuda13DirExists bool
-	cuda12DirExists bool
+	cuda13DirExists  bool
+	cuda12DirExists  bool
+	capabilityLogged bool
 )
 
 func init() {
@@ -133,7 +134,10 @@ func (s *SystemState) getSystemCapabilities() string {
 		return defaultCapability
 	}
 
-	xlog.Info("Capability automatically detected", "capability", s.GPUVendor, "env", capabilityEnv)
+	if !capabilityLogged {
+		xlog.Info("Capability automatically detected", "capability", s.GPUVendor, "env", capabilityEnv)
+		capabilityLogged = true
+	}
 	// If vram is less than 4GB, let's default to CPU but warn the user that they can override that via env
 	if s.VRAM <= 4*1024*1024*1024 {
 		xlog.Warn("VRAM is less than 4GB, defaulting to CPU", "env", capabilityEnv)