[NEW][koboldcpp][1.0.2]

koboldcpp/.helmignore

@@ -0,0 +1,23 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*.orig
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/

koboldcpp/Chart.yaml

@@ -0,0 +1,6 @@
+apiVersion: v2
+appVersion: "1.110"
+description: Run AI Models Locally, Free & Open-Source
+name: koboldcpp
+type: application
+version: 1.0.2

koboldcpp/OlaresManifest.yaml

@@ -0,0 +1,123 @@
+---
+olaresManifest.version: '0.11.0'
+olaresManifest.type: app
+apiVersion: 'v2'
+metadata:
+  name: koboldcpp
+  description: Run AI Models Locally, Free & Open-Source.
+  icon: https://app.cdn.olares.com/appstore/koboldcpp/icon.png
+  appid: koboldcpp
+  version: '1.0.2'
+  title: KoboldCpp
+  categories:
+  - Utilities_v112
+  - Productivity
+permission:
+  appData: true
+  appCache: true
+  userData:
+    - Home
+spec:
+  versionName: '1.110'
+  promoteImage:
+    - https://app.cdn.olares.com/appstore/koboldcpp/1.webp
+    - https://app.cdn.olares.com/appstore/koboldcpp/2.webp
+    - https://app.cdn.olares.com/appstore/koboldcpp/3.webp
+  fullDescription: |
+    *KoboldCPP is a powerful, C++ based backend built for running large language models locally using the GGUF format the same format supported by llama.cpp. Originally created to power storytelling and role-playing platforms like KoboldAI, it has grown into a complete local LLM engine capable of handling a wide variety of modern models, including:*
+    - LLaMA, LLaMA 2, and LLaMA 3
+    - Mistral and Mixtral
+    - Phi and Gemma
+    - Qwen and Yi
+    - Many other models converted to GGUF
+
+    *Features*
+    - Single file executable, with no installation required and no external dependencies
+    - Runs on CPU or GPU, supports full or partial offloaded
+    - LLM text generation (Supports all GGML and GGUF models, backwards compatibility with ALL past models)
+    - Image Generation and Image Editing (Stable Diffusion 1.5, SDXL, SD3, Flux, Qwen Image, Z-Image, Klein)
+    - Video Generation (WAN 2.2)
+    - Speech-To-Text (Voice Recognition) via Whisper
+    - Text-To-Speech (Voice Generation) via Qwen3TTS, Kokoro, OuteTTS, Parler and Dia
+    - Music Generation (Ace Step 1.5)
+    - Image Recognition (Multimodal Vision)
+    - MCP Server support and tool calling
+    - Provides many compatible APIs endpoints for many popular webservices (KoboldCppApi - OpenAiApi OllamaApi A1111ForgeApi ComfyUiApi WhisperTranscribeApi XttsApi OpenAiSpeechApi)
+    - Bundled KoboldAI Lite UI with editing tools, save formats, memory, world info, author's note, characters, scenarios.
+    - Includes multiple modes (chat, adventure, instruct, storywriter) and UI Themes (aesthetic roleplay, classic writer, corporate assistant, messsenger)
+    - Supports loading Tavern Character Cards, importing many different data formats from various sites, reading or exporting JSON savefiles and persistent stories.
+    - Many other features including new samplers, regex support, websearch, RAG via TextDB, image recognition/vision and more.
+    - Ready-to-use binaries for Windows, MacOS, Linux. Runs directly with Colab, Docker, also supports other platforms if self-compiled (like Android (via Termux) and Raspberry PI).
+  developer: KoboldAI
+  website: https://github.com/LostRuins/koboldcpp
+  submitter: Olares
+  locale:
+    - en-US
+    - zh-CN
+  doc: https://github.com/LostRuins/koboldcpp/wiki
+  {{- if and .Values.admin .Values.bfl.username (eq .Values.admin .Values.bfl.username) }}
+  requiredMemory: 4Gi
+  limitedMemory: 8Gi
+  requiredDisk: 5Gi
+  limitedDisk: 50Gi
+  requiredCpu: 2.1
+  limitedCpu: 6
+  requiredGpu: 10Gi
+  limitedGpu: 24Gi
+  {{- else }}
+  requiredMemory: 64Mi
+  limitedMemory: 500Mi
+  requiredDisk: 50Mi
+  limitedDisk: 200Mi
+  requiredCpu: 10m
+  limitedCpu: 500m
+  {{- end }}
+  supportArch:
+    - amd64
+  subCharts:
+    - name: koboldcppserver
+      shared: true
+    - name: koboldcpp
+options:
+  apiTimeout: 0
+  dependencies:
+    - name: olares
+      type: system
+      version: '>=1.12.3-0'
+  {{- if and .Values.admin .Values.bfl.username (eq .Values.admin .Values.bfl.username) }}
+  {{- else }}
+    - name: koboldcpp
+      type: application
+      version: '>=1.0.0'
+      mandatory: true
+  {{- end }}
+  appScope:
+  {{- if and .Values.admin .Values.bfl.username (eq .Values.admin .Values.bfl.username) }}
+    clusterScoped: true
+    appRef:
+      - koboldcpp
+  {{- else }}
+    clusterScoped: false
+  {{- end }}
+sharedEntrances:
+  - name: koboldcpp
+    host: sharedentrances-koboldcpp
+    port: 0
+    title: KoboldCpp API
+    icon: https://app.cdn.olares.com/appstore/koboldcpp/icon.png
+    invisible: true
+    authLevel: internal
+entrances:
+  - authLevel: internal
+    host: koboldcpp-web-svc
+    icon: https://app.cdn.olares.com/appstore/koboldcpp/icon.png
+    name: koboldcpp
+    openMethod: window
+    port: 8080
+    title: KoboldCpp
+envs:
+  - envName: OLARES_USER_HUGGINGFACE_SERVICE
+    required: true
+    applyOnChange: true
+    valueFrom:
+      envName: OLARES_USER_HUGGINGFACE_SERVICE

koboldcpp/i18n/en-US/OlaresManifest.yaml

@@ -0,0 +1,29 @@
+metadata:
+  title: KoboldCpp
+  description: Run AI Models Locally, Free & Open-Source.
+spec:
+  fullDescription: |
+    *KoboldCPP is a powerful, C++ based backend built for running large language models locally using the GGUF format the same format supported by llama.cpp. Originally created to power storytelling and role-playing platforms like KoboldAI, it has grown into a complete local LLM engine capable of handling a wide variety of modern models, including:*
+    - LLaMA, LLaMA 2, and LLaMA 3
+    - Mistral and Mixtral
+    - Phi and Gemma
+    - Qwen and Yi
+    - Many other models converted to GGUF
+
+    *Features*
+    - Single file executable, with no installation required and no external dependencies
+    - Runs on CPU or GPU, supports full or partial offloaded
+    - LLM text generation (Supports all GGML and GGUF models, backwards compatibility with ALL past models)
+    - Image Generation and Image Editing (Stable Diffusion 1.5, SDXL, SD3, Flux, Qwen Image, Z-Image, Klein)
+    - Video Generation (WAN 2.2)
+    - Speech-To-Text (Voice Recognition) via Whisper
+    - Text-To-Speech (Voice Generation) via Qwen3TTS, Kokoro, OuteTTS, Parler and Dia
+    - Music Generation (Ace Step 1.5)
+    - Image Recognition (Multimodal Vision)
+    - MCP Server support and tool calling
+    - Provides many compatible APIs endpoints for many popular webservices (KoboldCppApi - OpenAiApi OllamaApi A1111ForgeApi ComfyUiApi WhisperTranscribeApi XttsApi OpenAiSpeechApi)
+    - Bundled KoboldAI Lite UI with editing tools, save formats, memory, world info, author's note, characters, scenarios.
+    - Includes multiple modes (chat, adventure, instruct, storywriter) and UI Themes (aesthetic roleplay, classic writer, corporate assistant, messsenger)
+    - Supports loading Tavern Character Cards, importing many different data formats from various sites, reading or exporting JSON savefiles and persistent stories.
+    - Many other features including new samplers, regex support, websearch, RAG via TextDB, image recognition/vision and more.
+    - Ready-to-use binaries for Windows, MacOS, Linux. Runs directly with Colab, Docker, also supports other platforms if self-compiled (like Android (via Termux) and Raspberry PI).

koboldcpp/i18n/zh-CN/OlaresManifest.yaml

@@ -0,0 +1,29 @@
+metadata:
+  title: KoboldCpp
+  description: 本地运行 AI 模型，免费且开源。
+spec:
+  fullDescription: |
+    *KoboldCPP 是一个强大的基于 C++ 的后端，专为本地运行大型语言模型而设计，采用与 llama.cpp 相同支持的 GGUF 格式。最初为像 KoboldAI 这样的故事创作和角色扮演平台打造，如今已发展为一款功能全面的本地大模型引擎，能够支持多种现代模型，包括：*
+    - LLaMA、LLaMA 2 和 LLaMA 3
+    - Mistral 和 Mixtral
+    - Phi 和 Gemma
+    - Qwen 和 Yi
+    - 以及许多其他已转换为 GGUF 格式的模型
+
+    *功能特色*
+    - 单文件可执行程序，无需安装，无需任何外部依赖
+    - 支持 CPU 或 GPU 运行，可全量或部分转存
+    - LLM 文本生成（支持所有 GGML 和 GGUF 模型，同时向下兼容所有旧模型）
+    - 图像生成与编辑（支持 Stable Diffusion 1.5、SDXL、SD3、Flux、Qwen Image、Z-Image、Klein）
+    - 视频生成（WAN 2.2）
+    - 语音转文字（Whisper 语音识别）
+    - 文字转语音（通过 Qwen3TTS、Kokoro、OuteTTS、Parler 和 Dia 语音生成）
+    - 音乐生成（Ace Step 1.5）
+    - 图像识别（多模态视觉）
+    - 支持 MCP 服务器及工具调用
+    - 提供众多流行 Web 服务兼容的 API 端点（KoboldCppApi、OpenAiApi、OllamaApi、A1111ForgeApi、ComfyUiApi、WhisperTranscribeApi、XttsApi、OpenAiSpeechApi）
+    - 捆绑 KoboldAI Lite UI，内置编辑工具、存档格式、记忆功能、世界信息、作者笔记、角色、场景管理等
+    - 包括多种模式（聊天、冒险、指令、故事创作）和多种 UI 主题（美学角色扮演、经典作家、商务助手、信息使者）
+    - 支持加载 Tavern 角色卡，导入各类站点数据格式，读取/导出 JSON 存档与持久化故事
+    - 还包含新采样器、正则表达式支持、网页搜索、基于 TextDB 的 RAG、图像识别/视觉等众多功能
+    - 提供适用于 Windows、MacOS、Linux 的即用型二进制文件；支持 Colab、Docker 直接运行，自行编译还支持其他平台（如 Android（Termux）和树莓派）

koboldcpp/koboldcpp/Chart.yaml

@@ -0,0 +1,6 @@
+apiVersion: v2
+appVersion: "latest"
+description: description
+name: koboldcpp
+type: application
+version: 1.0.0

koboldcpp/koboldcpp/templates/clientproxy.yaml

@@ -0,0 +1,120 @@
+---
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: nginx-config
+  namespace: {{ .Release.Namespace }}
+data:
+  nginx.conf: |
+    upstream app_backend {
+      server koboldcpp-svc.koboldcppserver-shared:5001 max_fails=1 fail_timeout=2s;
+      server download-svc.koboldcppserver-shared:8090 backup;
+    }
+
+    server {
+      listen 8080;
+      access_log /opt/bitnami/openresty/nginx/logs/access.log;
+      error_log /opt/bitnami/openresty/nginx/logs/error.log;
+
+      client_max_body_size 200m;
+
+      location / {
+        proxy_pass http://app_backend;
+        proxy_next_upstream error timeout http_500 http_502 http_503 http_504;
+        proxy_next_upstream_tries 2;
+        proxy_connect_timeout 2s;
+        proxy_read_timeout 600s;
+        proxy_send_timeout 600s;
+
+        proxy_http_version 1.1;
+        proxy_set_header Host $host;
+        proxy_set_header X-Forwarded-Host $http_host;
+        proxy_set_header X-Real-IP $remote_addr;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+        proxy_set_header Upgrade $http_upgrade;
+        proxy_set_header Connection "upgrade";
+
+        client_max_body_size 200m;
+      }
+    }
+
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: {{ .Release.Name }}
+  namespace: {{ .Release.Namespace }}
+  labels:
+    io.kompose.service: koboldcppweb
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      io.kompose.service: koboldcppweb
+  strategy:
+    type: Recreate
+  template:
+    metadata:
+      labels:
+        io.kompose.service: koboldcppweb
+    spec:
+      volumes:
+        - name: nginx-config
+          configMap:
+            name: nginx-config
+            defaultMode: 438
+            items:
+              - key: nginx.conf
+                path: nginx.conf
+      containers:
+        - name: nginx
+          image: docker.io/beclab/aboveos-bitnami-openresty:1.25.3-2
+          ports:
+            - containerPort: 8080
+              protocol: TCP
+          env:
+            - name: OPENRESTY_CONF_FILE
+              value: /etc/nginx/nginx.conf
+          readinessProbe:
+            exec:
+              command:
+                - /bin/sh
+                - -c
+                - |
+                  http_code=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:8080)
+                  [ $http_code -ge 200 ] && [ $http_code -lt 500 ]
+            initialDelaySeconds: 2
+            timeoutSeconds: 3
+            periodSeconds: 3
+            successThreshold: 1
+            failureThreshold: 60
+          resources:
+            limits:
+              cpu: 100m
+              memory: 256Mi
+            requests:
+              cpu: 10m
+              memory: 64Mi
+          volumeMounts:
+            - name: nginx-config
+              mountPath: /etc/nginx/nginx.conf
+              subPath: nginx.conf
+            - name: nginx-config
+              mountPath: /opt/bitnami/openresty/nginx/conf/server_blocks/nginx.conf
+              subPath: nginx.conf
+
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: koboldcpp-web-svc
+  namespace: {{ .Release.Namespace }}
+spec:
+  type: ClusterIP
+  selector:
+    io.kompose.service: koboldcppweb
+  ports:
+    - name: http
+      protocol: TCP
+      port: 8080
+      targetPort: 8080

koboldcpp/koboldcppserver/Chart.yaml

@@ -0,0 +1,6 @@
+apiVersion: v2
+appVersion: "latest"
+description: description
+name: koboldcppserver
+type: application
+version: 1.0.0

koboldcpp/koboldcppserver/templates/configmap.yaml

@@ -0,0 +1,13 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: koboldcpp-env
+  namespace: {{ .Release.Namespace }}
+data:
+  NVIDIA_VISIBLE_DEVICES: "all"
+  NVIDIA_DRIVER_CAPABILITIES: "compute,utility"
+  KCPP_MODEL: ""
+  KCPP_ARGS: "--model /models/Qwen3.5-4B-UD-Q4_K_XL.gguf --contextsize 4096 --usecuda --gpulayers 99 --mmproj /models/mmproj-F32.gguf --sdmodel /models/picX_real.safetensors --sdquant 1 --whispermodel /models/ggml-large-v3-turbo.bin --ttsmodel /models/Qwen3-TTS-12Hz-1.7B-Base-q8_0.gguf --ttswavtokenizer /models/qwen3-tts-tokenizer-q8_0.gguf --ttsgpu --admin --admindir /models/admindir"
+  KCPP_DONT_UPDATE: "true"
+  KCPP_DONT_TUNNEL: "true"
+  KCPP_DONT_REMOVE_MODELS: "true"