feat: aichat service integration, log levels, ollama modelfiles

README.md

@@ -152,7 +152,7 @@ harbor open
 
 ##### Satellites
 
-[SearXNG](https://github.com/av/harbor/wiki/Services#searxng) ⦁︎ [Perplexica](https://github.com/av/harbor/wiki/Services#perplexica) ⦁︎ [Dify](https://github.com/av/harbor/wiki/Services#dify) ⦁︎ [Plandex](https://github.com/av/harbor/wiki/Services#plandex) ⦁︎ [LiteLLM](https://github.com/av/harbor/wiki/Services#-litellm) ⦁︎ [LangFuse](https://github.com/av/harbor/wiki/Services#langfuse) ⦁︎ [Open Interpreter](https://github.com/av/harbor/wiki/Services#-open-interpreter) ⦁︎ [cloudflared](https://github.com/av/harbor/wiki/Services#cloudflared) ⦁︎ [cmdh](https://github.com/av/harbor/wiki/Services#cmdh) ⦁︎ [fabric](https://github.com/av/harbor/wiki/Services#fabric) ⦁︎ [txtai RAG](https://github.com/av/harbor/wiki/Services#txtai-rag) ⦁︎ [TextGrad](https://github.com/av/harbor/wiki/Services#textgrad) ⦁︎ [Aider](https://github.com/av/harbor/wiki/Services#aider)
+[SearXNG](https://github.com/av/harbor/wiki/Services#searxng) ⦁︎ [Perplexica](https://github.com/av/harbor/wiki/Services#perplexica) ⦁︎ [Dify](https://github.com/av/harbor/wiki/Services#dify) ⦁︎ [Plandex](https://github.com/av/harbor/wiki/Services#plandex) ⦁︎ [LiteLLM](https://github.com/av/harbor/wiki/Services#-litellm) ⦁︎ [LangFuse](https://github.com/av/harbor/wiki/Services#langfuse) ⦁︎ [Open Interpreter](https://github.com/av/harbor/wiki/Services#-open-interpreter) ⦁︎ [cloudflared](https://github.com/av/harbor/wiki/Services#cloudflared) ⦁︎ [cmdh](https://github.com/av/harbor/wiki/Services#cmdh) ⦁︎ [fabric](https://github.com/av/harbor/wiki/Services#fabric) ⦁︎ [txtai RAG](https://github.com/av/harbor/wiki/Services#txtai-rag) ⦁︎ [TextGrad](https://github.com/av/harbor/wiki/Services#textgrad) ⦁︎ [Aider](https://github.com/av/harbor/wiki/Services#aider) ⦁︎ [aichat](https://github.com/av/harbor/wiki/Services#aichat)
 
 ## Why?
 

aichat/Dockerfile

@@ -0,0 +1,10 @@
+FROM python:3.11
+SHELL ["/bin/bash", "-c"]
+
+WORKDIR /app
+RUN pip install pyyaml
+
+RUN curl https://zyedidia.github.io/eget.sh | sh
+RUN ./eget sigoden/aichat
+
+ENTRYPOINT [ "/app/aichat" ]

aichat/configs/aichat.airllm.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: airllm
+    api_base: http://airllm:5000/v1
+    api_key: sk-airllm
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/configs/aichat.aphrodite.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: aphrodite
+    api_base: http://aphrodite:7860/v1
+    api_key: sk-aphrodite
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/configs/aichat.config.yml

@@ -0,0 +1,115 @@
+# This is where other configs will be merged into
+# See official example: https://github.com/sigoden/aichat/blob/main/config.example.yaml
+
+# ---- llm ----
+
+# This field requires a prefix matching
+# the name of the openai-compatible backend,
+# so it's set from the "cross" files, e.g. aichat.ollama.yml
+# model: ollama:${HARBOR_AICHAT_MODEL}
+
+temperature: null                # Set default temperature parameter
+top_p: null                      # Set default top-p parameter, range (0, 1)
+
+# ---- behavior ----
+stream: true                     # Controls whether to use the stream-style API.
+save: true                       # Indicates whether to persist the message
+keybindings: emacs               # Choose keybinding style (emacs, vi)
+editor: null                     # Specifies the command used to edit input buffer or session. (e.g. vim, emacs, nano).
+wrap: 'auto'                     # Controls text wrapping (no, auto, <max-width>)
+wrap_code: false                 # Enables or disables wrapping of code blocks
+
+# ---- prelude ----
+prelude: null                    # Set a default role or session to start with (e.g. role:<name>, session:<name>)
+repl_prelude: null               # Overrides the `prelude` setting specifically for conversations started in REPL
+agent_prelude: null              # Set a session to use when starting a agent. (e.g. temp, default)
+
+# ---- session ----
+# Controls the persistence of the session. if true, auto save; if false, not save; if null, asking the user
+save_session: null
+# Compress session when token count reaches or exceeds this threshold
+compress_threshold: 4000
+# Text prompt used for creating a concise summary of session message
+summarize_prompt: 'Summarize the discussion briefly in 200 words or less to use as a prompt for future context.'
+# Text prompt used for including the summary of the entire session
+summary_prompt: 'This is a summary of the chat history as a recap: '
+
+# ---- function-calling ----
+# Visit https://github.com/sigoden/llm-functions for setup instructions
+function_calling: true           # Enables or disables function calling (Globally).
+mapping_tools:                   # Alias for a tool or toolset
+  fs: 'fs_cat,fs_ls,fs_mkdir,fs_rm,fs_write'
+use_tools: null                  # Which tools to use by default. (e.g. 'fs,web_search')
+
+# ---- RAG ----
+# See [RAG-Guide](https://github.com/sigoden/aichat/wiki/RAG-Guide) for more details.
+rag_embedding_model: null                   # Specifies the embedding model to use
+rag_reranker_model: null                    # Specifies the rerank model to use
+rag_top_k: 4                                # Specifies the number of documents to retrieve
+rag_chunk_size: null                        # Specifies the chunk size
+rag_chunk_overlap: null                     # Specifies the chunk overlap
+rag_min_score_vector_search: 0              # Specifies the minimum relevance score for vector-based searching
+rag_min_score_keyword_search: 0             # Specifies the minimum relevance score for keyword-based searching
+rag_min_score_rerank: 0                     # Specifies the minimum relevance score for reranking
+# Defines the query structure using variables like __CONTEXT__ and __INPUT__ to tailor searches to specific needs
+rag_template: |
+  Use the following context as your learned knowledge, inside <context></context> XML tags.
+  <context>
+  __CONTEXT__
+  </context>
+
+  When answer to user:
+  - If you don't know, just say that you don't know.
+  - If you don't know when you are not sure, ask for clarification.
+  Avoid mentioning that you obtained the information from the context.
+  And answer according to the language of the user's question.
+
+  Given the context information, answer the query.
+  Query: __INPUT__
+# Define document loaders to control how RAG and `.file`/`--file` load files of specific formats.
+document_loaders:
+  # You can add custom loaders using the following syntax:
+  #   <file-extension>: <command-to-load-the-file>
+  # Note: Use `$1` for input file and `$2` for output file. If `$2` is omitted, use stdout as output.
+  pdf: 'pdftotext $1 -'                         # Load .pdf file, see https://poppler.freedesktop.org to set up pdftotext
+  docx: 'pandoc --to plain $1'                  # Load .docx file, see https://pandoc.org to set up pandoc
+
+# ---- appearance ----
+highlight: true                  # Controls syntax highlighting
+light_theme: false               # Activates a light color theme when true. env: AICHAT_LIGHT_THEME
+# Custom REPL left/right prompts, see https://github.com/sigoden/aichat/wiki/Custom-REPL-Prompt for more details
+left_prompt:
+  '{color.green}{?session {?agent {agent}>}{session}{?role /}}{!session {?agent {agent}>}}{role}{?rag @{rag}}{color.cyan}{?session )}{!session >}{color.reset} '
+right_prompt:
+  '{color.purple}{?session {?consume_tokens {consume_tokens}({consume_percent}%)}{!consume_tokens {consume_tokens}}}{color.reset}'
+
+# ---- clients ----
+clients: []
+  # All clients have the following configuration:
+  # - type: xxxx
+  #   name: xxxx                                      # Only use it to distinguish clients with the same client type. Optional
+  #   models:
+  #     - name: xxxx                                  # Chat model
+  #       max_input_tokens: 100000
+  #       supports_vision: true
+  #       supports_function_calling: true
+  #     - name: xxxx                                  # Embedding model
+  #       type: embedding
+  #       max_input_tokens: 2048
+  #       max_tokens_per_chunk: 2048
+  #       default_chunk_size: 1500
+  #       max_batch_size: 100
+  #     - name: xxxx                                  # Reranker model
+  #       type: reranker
+  #       max_input_tokens: 2048
+  #   patch:                                          # Patch api
+  #     chat_completions:                             # Api type, possible values: chat_completions, embeddings, and rerank
+  #       <regex>:                                    # The regex to match model names, e.g. '.*' 'gpt-4o' 'gpt-4o|gpt-4-.*'
+  #         url: ''                                   # Patch request url
+  #         body:                                     # Patch request body
+  #           <json>
+  #         headers:                                  # Patch request headers
+  #           <key>: <value>
+  #   extra:
+  #     proxy: socks5://127.0.0.1:1080                # Set proxy
+  #     connect_timeout: 10                           # Set timeout in seconds for connect to api

aichat/configs/aichat.dify.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: dify
+    api_base: http://dify-openai:3000/v1
+    apiKey: "${HARBOR_DIFY_OPENAI_WORKFLOW}"
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/configs/aichat.litellm.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: litellm
+    api_base: http://litellm:3000/v1
+    apiKey: "${HARBOR_LITELLM_MASTER_KEY}"
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/configs/aichat.llamacpp.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: llamacpp
+    api_base: http://llamacpp:8080/v1
+    apiKey: sk-llamacpp
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/configs/aichat.mistralrs.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: mistralrs
+    api_base: http://mistralrs:8021/v1
+    apiKey: sk-mistralrs
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/configs/aichat.ollama.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: ollama
+    api_base: http://ollama:11434/v1
+    api_key: sk-ollama
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/configs/aichat.tabbyapi.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: tabbyapi
+    api_base: http://tabbyapi:5000/v1
+    apiKey: "${HARBOR_TABBYAPI_ADMIN_KEY}"
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/configs/aichat.vllm.yml

@@ -0,0 +1,8 @@
+clients:
+  - type: openai-compatible
+    name: tabbyapi
+    api_base: http://vllm:8000/v1
+    apiKey: sk-vllm
+    models:
+      - name: ${HARBOR_AICHAT_MODEL}
+

aichat/start_aichat.sh

@@ -0,0 +1,19 @@
+#!/bin/bash
+
+log() {
+  if [ "$HARBOR_LOG_LEVEL" == "DEBUG" ]; then
+    echo "$1"
+  fi
+}
+
+log "Harbor: custom aichat entrypoint"
+
+log "YAML Merger is starting..."
+mkdir -p /root/.config/aichat
+python /app/yaml_config_merger.py --pattern ".yml" --output "/root/.config/aichat/config.yaml" --directory "/app/configs"
+
+log "Merged Configs:"
+log $(cat /root/.config/aichat/config.yaml)
+
+log echo "Starting aichat with args: '$*'"
+/app/aichat $@

compose.aichat.yml

@@ -0,0 +1,20 @@
+services:
+  aichat:
+    container_name: ${HARBOR_CONTAINER_PREFIX}.aichat
+    build:
+      context: ./aichat
+      dockerfile: Dockerfile
+    env_file:
+      - ./.env
+    networks:
+      - harbor-network
+    volumes:
+      # Base config
+      - ./aichat/configs/aichat.config.yml:/app/configs/z.config.yml
+      # Custom entrypoint for config merging
+      - ./aichat/start_aichat.sh:/app/start_aichat.sh
+      - ./shared/yaml_config_merger.py:/app/yaml_config_merger.py
+      - ${HARBOR_AICHAT_CONFIG_PATH}:/root/.config/aichat
+    ports:
+      - ${HARBOR_AICHAT_HOST_PORT}:${HARBOR_AICHAT_HOST_PORT}
+    entrypoint: ["/app/start_aichat.sh"]

compose.ollama.yml

@@ -4,6 +4,7 @@ services:
     env_file: ./.env
     volumes:
       - ${HARBOR_OLLAMA_CACHE}:/root/.ollama
+      - ./ollama/modelfiles:/modelfiles
     tty: true
     image: ollama/ollama:${HARBOR_OLLAMA_VERSION}
     ports:

compose.x.aichat.ollama.yml

@@ -0,0 +1,4 @@
+services:
+  aichat:
+    volumes:
+      - ./aichat/configs/aichat.ollama.yml:/app/configs/ollama.yml

default.env

@@ -56,6 +56,7 @@ HARBOR_CONTAINER_PREFIX="harbor"
 HARBOR_CLI_NAME="harbor"
 HARBOR_CLI_SHORT="h"
 HARBOR_CLI_PATH="~/.local/bin"
+HARBOR_LOG_LEVEL="INFO"
 
 # OpenAI
 # ---------------------
@@ -246,6 +247,11 @@ HARBOR_COMFYUI_PROVISIONING="https://raw.githubusercontent.com/av/harbor/main/co
 HARBOR_PERPLEXICA_HOST_PORT=34041
 HARBOR_PERPLEXICA_BACKEND_HOST_PORT=34042
 
+# Aichat
+HARBOR_AICHAT_HOST_PORT=34051
+HARBOR_AICHAT_MODEL="llama3.1:8b"
+HARBOR_AICHAT_CONFIG_PATH="~/.config/aichat"
+
 # ============================================
 # Service Configuration.
 # You can specify any of the service's own environment variables here.