Ollama-Copilot - GitLite

KO | EN

gitlite — search

#javascript #python #react #ai #llm #go #golang #machine-learning #deep-learning #macos #windows #vue #security #openai

Ollama-Copilot

(★ 87)

Neovim plugin for code completion

File Explorer

Download Latest Version (.zip)

- demoV1.gif
- demoV2.gif
- - ghost_text.lua
  - init.lua
  - lsp_client.lua
  - start_up.lua
- benchmark.py
- embeddings.py
- generate.py
- Modelfile
- ollama_rest.py
- ollama-copilot.lua
- completion_engine.py
- ollama_lsp.py
- requestmanager.py
- requirements.txt
.gitignore
ideas.md
ollama.md
README.md

# Installation Guide

1. Get the code

git clone https://github.com/Jacob411/Ollama-Copilot

Downloads the entire project code from GitHub to your computer.

cd Ollama-Copilot

Moves into the project folder you just downloaded.

2. Official Install Script

Easy Recommended

Prerequisites

Python 3 Python is required to use pip.

curl -fsSL https://ollama.com/install.sh | sh

Downloads and runs the official install script in one line — this handles the full setup automatically.

pip install pygls ollama

Installs the package published on PyPI directly — no need to clone the source.

✅ After installing, open a new terminal and run the program's version command (e.g. --version) to confirm it worked.

Pulled directly from this repo's README.

3. Python

Easy

Prerequisites

Git Needed to download the project code from GitHub.
Python 3 On Windows, be sure to check 'Add Python to PATH' during installation.

pip install pygls ollama

Installs the package published on PyPI directly — no need to clone the source.

✅ If it runs without errors and prints output in the terminal, it worked.

Pulled directly from this repo's README.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Ollama Copilot
<img src="assets/demoV2.gif" width="900"  alt="Demo GIF">

## Overview
### Copilot-like Tab Completion for NeoVim
Ollama Copilot allows users to integrate their Ollama code completion models into Neovim, giving GitHub Copilot-like tab completions.  
  
Offers **Suggestion Streaming** which will stream the completions into your editor as they are generated from the model.

### Optimizations:
- [x] Debouncing for subsequent completion requests to avoid overflows of Ollama requests which lead to CPU over-utilization.
- [x] Full control over triggers, using textChange events instead of Neovim client requests.
### Features
- [x] Language server which can provide code completions from an Ollama model
- [x] Ghost text completions which can be inserted into the editor
- [x] Streamed ghost text completions which populate in real-time

## Install
### Requires
To use Ollama-Copilot, you need to have Ollama installed [github.com/ollama/ollama](https://github.com/ollama/ollama):  
```bash
curl -fsSL https://ollama.com/install.sh | sh
```
Also, the language server runs on Python, and requires two libraries (Can also be found in python/requirements.txt)
```bash
pip install pygls ollama
```
Make sure you have the model you want to use installed, a catalog can be found here: [ollama.com/library](https://ollama.com/library?q=code)
```
# To view your available models:
ollama ls

# To pull a new model
ollama pull <Model name>
```
### Using a plugin manager
Lazy:
```lua
-- Default configuration
{"Jacob411/Ollama-Copilot", opts={}}
```
```lua
-- Custom configuration (defaults shown)
{
  'jacob411/Ollama-Copilot',
  opts = {
    model_name = "deepseek-coder:base",
    ollama_url = "http://localhost:11434", -- URL for Ollama server, Leave blank to use default local instance.
    stream_suggestion = false,
    python_command = "python3",
    filetypes = {'python', 'lua','vim', "markdown"},
    capabilities = nil, -- LSP capabilities, auto-detected if not provided
    ollama_model_opts = {
        num_predict = 40,
        temperature = 0.1,
    },
    keymaps = {
        suggestion = '<leader>os',
        reject = '<leader>or',
        insert_accept = '<Tab>',
    },
}
},
```
For more Ollama customization, see [github.com/ollama/ollama/blob/main/docs/modelfile.md](https://github.com/ollama/ollama/blob/main/docs/modelfile.md)

### LSP Capabilities Configuration

The plugin automatically detects and configures LSP capabilities for optimal completion support:

1. **Auto-detection (default)**: If `capabilities` is not specified, the plugin will:
   - Try to use `cmp_nvim_lsp.default_capabilities()` if nvim-cmp is installed
   - Fall back to `vim.lsp.protocol.make_client_capabilities()` if nvim-cmp is not available

2. **Custom capabilities**: You can override the auto-detection by providing your own capabilities:
   ```lua
   opts = {
     capabilities = require('cmp_nvim_lsp').default_capabilities(),
     -- or use custom capabilities
     capabilities = vim.tbl_deep_extend('force',
       vim.lsp.protocol.make_client_capabilities(),
       { your_custom_capability = true }
     )
   }
   ```

This ensures backward compatibility while allowing the plugin to work without requiring nvim-cmp as a dependency.

## Usage
Ollama copilot language server will attach when you enter a buffer and can be viewed using:
```lua
:LspInfo
```
### Recomendations
Smaller models (<3 billion parameters) work best for tab completion tasks, providing low latency and minimal CPU usage.
- [deepseek-coder](https://ollama.com/library/deepseek-coder:1.3b) - 1.3B
- [starcoder](https://ollama.com/library/starcoder:1b) - 1B
- [codegemma](https://ollama.com/library/codegemma:2b) - 2B
- [starcoder2](https://ollama.com/library/starcoder2:3b) - 3B
  
## Contributing
Contributions are welcome! If you have any ideas for new features, improvements, or bug fixes, please open an issue or submit a pull request.

I am hopeful to add more on the model side as well, as I am interested in finetuning the models and implementing RAG techniques, moving outside of using just Ollama.

## License
This project is licensed under the MIT License.