ollama

Dockerfile

@@ -1,39 +1,9 @@
-# Builder stage
-FROM ubuntu:latest
+FROM ollama/ollama
 
-# Update packages and install curl and gnupg
-RUN apt-get update && apt-get install -y \
-    curl \
-    gnupg
-
-# Add NVIDIA package repositories
-RUN curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
-    && echo "deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://nvidia.github.io/libnvidia-container/stable/deb/ $(. /etc/os-release; echo $UBUNTU_CODENAME) main" > /etc/apt/sources.list.d/nvidia-container-toolkit.list
-
-# Install NVIDIA container toolkit (Check for any updated methods or URLs for Ubuntu jammy)
-RUN apt-get update && apt-get install -y nvidia-container-toolkit || true
-
-# Install application
-RUN curl https://ollama.ai/install.sh | sh
-# Below is to fix embedding bug as per
-# RUN curl -fsSL https://ollama.com/install.sh | sed 's#https://ollama.com/download#https://github.com/jmorganca/ollama/releases/download/v0.1.29#' | sh
-
-
-# Create the directory and give appropriate permissions
 RUN mkdir -p /.ollama && chmod 777 /.ollama
 
-WORKDIR /.ollama
-
-# Copy the entry point script
-COPY entrypoint.sh /entrypoint.sh
-RUN chmod +x /entrypoint.sh
-
-# Set the entry point script as the default command
-ENTRYPOINT ["/entrypoint.sh"]
-CMD ["ollama", "serve"]
-
-# Set the model as an environment variable (this can be overridden)
-ENV model=${model}
+ENV OLLAMA_MAX_LOADED_MODELS=20 --OLLAMA_NUM_PARALLEL=20
+ENV OLLAMA_HOST "0.0.0.0:7860"
 
-# Expose the server port
-EXPOSE 7860
+CMD ["serve"]
+EXPOSE 7860

app.py

@@ -1,10 +0,0 @@
-from langchain_community.llms import Ollama
-import gradio as gr
-
-def generate(user_input):
-    llm = Ollama(model="openchat")
-    response = llm.invoke(user_input)
-    return response
-
-iface = gr.Interface(fn=generate, inputs="text", outputs="text", title="Ollama Chat", description="Chat with Ollama by entering your message.")
-iface.launch()

entrypoint.sh

@@ -1,21 +0,0 @@
-#!/bin/bash
-
-# Starting server
-echo "Starting server"
-ollama serve &
-sleep 1
-
-# Splitting the models by comma and pulling each
-IFS=',' read -ra MODELS <<< "$model"
-for m in "${MODELS[@]}"; do
-    echo "Pulling $m"
-    ollama pull "$m"
-    sleep 5
-    # echo "Running $m"
-    # ollama run "$m"
-    # No need to sleep here unless you want to give some delay between each pull for some reason
-    python3 app.py
-done
-
-# Keep the script running to prevent the container from exiting
-wait

ollama-api-demo.ipynb

@@ -1,220 +0,0 @@
-{
- "cells": [
-  {
-   "attachments": {},
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Dependencies"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "%pip install openai --upgrade"
-   ]
-  },
-  {
-   "attachments": {},
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## API Response"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 68,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:42:59.839736985Z\",\"response\":\"```\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:42:59.859007873Z\",\"response\":\"\\n\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:42:59.878431213Z\",\"response\":\"def\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:42:59.897784641Z\",\"response\":\" add\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:42:59.91718876Z\",\"response\":\"(\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:42:59.936866527Z\",\"response\":\"a\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:42:59.95776024Z\",\"response\":\",\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:42:59.979133947Z\",\"response\":\" b\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.000494731Z\",\"response\":\"):\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.021318934Z\",\"response\":\"\\n\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.041779731Z\",\"response\":\"   \",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.062190588Z\",\"response\":\" return\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.082505875Z\",\"response\":\" a\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.102662719Z\",\"response\":\" +\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.122760355Z\",\"response\":\" b\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.142907745Z\",\"response\":\"\\n\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.163285108Z\",\"response\":\"```\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.18370624Z\",\"response\":\"\\n\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.203963933Z\",\"response\":\"Example\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.224025854Z\",\"response\":\" usage\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.244386112Z\",\"response\":\":\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.264846213Z\",\"response\":\"\\n\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.285448321Z\",\"response\":\"```\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.305657169Z\",\"response\":\"\\n\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.325782131Z\",\"response\":\"print\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.346353022Z\",\"response\":\"(\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.366430166Z\",\"response\":\"add\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.386881006Z\",\"response\":\"(\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.406680624Z\",\"response\":\"3\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.426827031Z\",\"response\":\",\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.447157302Z\",\"response\":\" \",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.467234406Z\",\"response\":\"5\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.487442969Z\",\"response\":\"))\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.50753674Z\",\"response\":\" #\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.527739408Z\",\"response\":\" Output\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.54789446Z\",\"response\":\":\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.568672362Z\",\"response\":\" \",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.591076535Z\",\"response\":\"8\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.614757129Z\",\"response\":\"\\n\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.637841098Z\",\"response\":\"```\",\"done\":false}\n",
-      "{\"model\":\"codellama\",\"created_at\":\"2024-03-20T17:43:00.660407109Z\",\"response\":\"\",\"done\":true,\"context\":[518,25580,29962,3532,14816,29903,29958,5299,829,14816,29903,6778,13,13,6113,5132,775,304,788,29871,29906,3694,518,29914,25580,29962,13,28956,13,1753,788,29898,29874,29892,289,1125,13,1678,736,263,718,289,13,28956,13,14023,8744,29901,13,28956,13,2158,29898,1202,29898,29941,29892,29871,29945,876,396,10604,29901,29871,29947,13,28956],\"total_duration\":10037918982,\"load_duration\":9097178085,\"prompt_eval_count\":28,\"prompt_eval_duration\":119308000,\"eval_count\":41,\"eval_duration\":820449000}\n"
-     ]
-    }
-   ],
-   "source": [
-    "!curl https://thewise-ollama-server.hf.space/api/generate -d '''{\"model\": \"codellama\",\"prompt\":\"Write Python code to add 2 numbers\"}'''"
-   ]
-  },
-  {
-   "attachments": {},
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Langchain Demo"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 69,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from langchain.llms import Ollama\n",
-    "from langchain.callbacks.manager import CallbackManager\n",
-    "from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler"
-   ]
-  },
-  {
-   "attachments": {},
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "##### CODELLAMA"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 70,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "```\n",
-      "def add(a, b):\n",
-      "    return a + b\n",
-      "```\n",
-      "This function takes two arguments `a` and `b`, adds them together, and returns the result. You can call this function by passing in two numbers, like this:\n",
-      "```\n",
-      "print(add(3, 5)) # prints 8\n",
-      "```"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "'```\\ndef add(a, b):\\n    return a + b\\n```\\nThis function takes two arguments `a` and `b`, adds them together, and returns the result. You can call this function by passing in two numbers, like this:\\n```\\nprint(add(3, 5)) # prints 8\\n```'"
-      ]
-     },
-     "execution_count": 70,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "llm = Ollama(\n",
-    "    model=\"codellama\",\n",
-    "    base_url=\"https://thewise-ollama-server.hf.space\",\n",
-    "    callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]))\n",
-    "\n",
-    "llm('Write Python code to add 2 numbers')"
-   ]
-  },
-  {
-   "attachments": {},
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "##### LLAMA2"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 71,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "```\n",
-      "# Adding two numbers\n",
-      "a = 5\n",
-      "b = 3\n",
-      "result = a + b\n",
-      "print(result) # Output: 8\n",
-      "```"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "'```\\n# Adding two numbers\\na = 5\\nb = 3\\nresult = a + b\\nprint(result) # Output: 8\\n```'"
-      ]
-     },
-     "execution_count": 71,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "llm = Ollama(\n",
-    "    model=\"llama2\",\n",
-    "    base_url=\"https://thewise-ollama-server.hf.space\",\n",
-    "    callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]))\n",
-    "\n",
-    "llm('Write Python code to add 2 numbers')"
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "langchain",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.11.4"
-  },
-  "orig_nbformat": 4
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}