Mochi Coco 🍡
.-===-.
|[:::]|
`-----´
A beautiful, feature-rich CLI chat application for interacting with LLMs via Ollama with streaming responses, session persistence, markdown rendering and tool support.
Installation
pip install mochi-coco
Quick Start
- Make sure you have Ollama running locally
- Pull a model:
ollama pull gpt-oss:20b - Start chatting:
mochi-coco
- Or directly open a specific chat session:
mochi-coco --chat 1
Features
- 🚀 Streaming responses - Real-time chat with immediate feedback
- ⚡ Stream interruption - Press ESC during streaming to interrupt and save partial responses
- 💾 Session persistence - Your conversations are automatically saved in the terminal's directory and resumable
- 🎨 Rich markdown rendering - Beautiful formatting with syntax highlighting and toggle rendering mid session
- 🔄 Model switching - Change models mid-conversation
- ✏️ Message editing - Edit previous messages and start from there
- 🧠 Thinking blocks - Toggle display of model reasoning (when supported by model + only in markdown mode)
- 📋 Session management - Switch between different chat sessions
- 🎛️ Interactive menus - Easy-to-use command interface with clear instructions
- ⚡ Background summarization - Automatic conversation summaries
- 📝 System Prompts - Drop
*.mdor*.txtfiles into thesystem_promptsfolder in the root directory of the terminal to use as system prompts. - Tool support - Let the LLM use tools to you've created by placing them in the
toolsfolder in the root directory of the terminal.
Session Commands
While chatting, you can use these commands:
/menu- Open the main menu with all options OR type in the following shortcuts:/chats- Switch between existing sessions or create new ones/models- Change the current model/markdown- Toggle markdown rendering on/off/thinking- Toggle thinking blocks display (only when markdown rendering is enabled)/system- Change system prompt during chat session/tools- Enable/disable tools feature (only when/toolsfolder exists in the terminal's root directory)
/status- Show status information about the current session (e.g. model, tools enabled/disabled, context window size, etc.)/edit- Edit a previous message and continue from there/exitor/quit- Exit the application
Usage
Basic Chat
mochi-coco
Custom Ollama Host
mochi-coco --host http://localhost:11434
Example Session
$ mochi-coco
╭─────────────────────────────────────── 🍡 Welcome to Mochi-Coco! ───────────────────────────────────────╮
│ │
│ 🤖 AI Chat with Style | v0.4.0 │
│ │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭───────────────────────────────────────── 💬 Previous Sessions ──────────────────────────────────────────╮
│ │
│ ┏━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓ │
│ ┃ # ┃ Session ID ┃ Model ┃ Preview ┃ Messages ┃ │
│ ┡━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩ │
│ │ 1 │ bc8a8d5c8e │ gpt-oss:20b │ Who was the first Avenger in the │ 26 │ │
│ │ │ │ │ MCU and who th... │ │ │
│ └─────┴──────────────┴──────────────────────┴─────────────────────────────────────┴──────────┘ │
│ │
│ 💡 Options: │
│ • 📝 Select session (1-1) │
│ • 🆕 Type 'new' for new chat │
│ • 🗑️ Type '/delete <number>' to delete session │
│ • 👋 Type 'q' to quit │
│ │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice: new
╭─ 🤖 Available Models ───────────────────────────────────────────────────────────────────────────────────╮
│ ╭─────┬───────────────────────────┬──────────────┬─────────────────┬──────────┬───────╮ │
│ │ # │ Model Name │ Size (MB) │ Family │ Max. Cxt │ Tools │ │
│ ├─────┼───────────────────────────┼──────────────┼─────────────────┼──────────┼───────┤ │
│ │ 1 │ qwen3-vl:32b │ 19941.6 │ qwen3vl │ 262144 │ Yes │ │
│ │ 2 │ qwen3-coder:latest │ 17697.0 │ qwen3moe │ 262144 │ Yes │ │
│ │ 3 │ gpt-oss:20b │ 13141.8 │ gptoss │ 131072 │ Yes │ │
│ │ 4 │ qwen3:14b │ 8846.5 │ qwen3 │ 40960 │ Yes │ │
│ │ 5 │ qwen3:latest │ 4983.3 │ qwen3 │ 40960 │ Yes │ │
│ │ 6 │ qwen3:30b │ 17697.0 │ qwen3moe │ 262144 │ Yes │ │
│ │ 7 │ llama3.2:latest │ 1925.8 │ llama │ 131072 │ Yes │ │
│ │ 8 │ mistral-small3.2:latest │ 14474.3 │ mistral3 │ 131072 │ Yes │ │
│ ╰─────┴───────────────────────────┴──────────────┴─────────────────┴──────────┴───────╯ │
│ │
│ 💡 Options: │
│ • 🔢 Select model (1-8) │
│ • 👋 Type 'q' to quit │
│ │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice: 3
╭────────────────────────────────────────── Markdown Rendering ───────────────────────────────────────────╮
│ 📝 Enable markdown formatting for responses? │
│ This will format code blocks, headers, tables, etc. │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Enable markdown? [y/n] (y): y
╭──────────────────────────────────────── Thinking Block Display ─────────────────────────────────────────╮
│ 🤔 Show model's thinking process in responses? │
│ This will display thinking blocks as formatted quotes. │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Show thinking blocks? [y/n] (n): y
╭─ 🔧 System Prompts ─────────────────────────────────────────────────────────────────────────────────────╮
│ ╭─────┬─────────────────┬─────────────────────────────────────┬────────────╮ │
│ │ # │ Filename │ Preview │ Word Count │ │
│ ├─────┼─────────────────┼─────────────────────────────────────┼────────────┤ │
│ │ 1 │ AGENTS.md │ You are a helpful agent │ 5 │ │
│ ╰─────┴─────────────────┴─────────────────────────────────────┴────────────╯ │
│ │
│ 💡 Options: │
│ • 📝 Select system prompt (1-1) │
│ • 🆕 Type 'no' for no system prompt │
│ • 🗑️ Type '/delete <number>' to delete a system prompt │
│ • 👋 Type 'q' to quit │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice: no
╭─────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ 💡 System prompt removed. │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─ Success ───────────────────────────────────────────────────────────────────────────────────────────────╮
│ Switched to session a102ba9d9a with gpt-oss:20b │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─ Info ──────────────────────────────────────────────────────────────────────────────────────────────────╮
│ The current model 'gpt-oss:20b' doesn't support structured summarization. Please select a compatible │
│ model to use for generating conversation summaries. │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─ 🤖 Available Models ───────────────────────────────────────────────────────────────────────────────────╮
│ ╭─────┬───────────────────────────┬──────────────┬─────────────────┬──────────┬───────╮ │
│ │ # │ Model Name │ Size (MB) │ Family │ Max. Cxt │ Tools │ │
│ ├─────┼───────────────────────────┼──────────────┼─────────────────┼──────────┼───────┤ │
│ │ 1 │ qwen3-vl:32b │ 19941.6 │ qwen3vl │ 262144 │ Yes │ │
│ │ 2 │ qwen3-coder:latest │ 17697.0 │ qwen3moe │ 262144 │ Yes │ │
│ │ 3 │ qwen3:latest │ 4983.3 │ qwen3 │ 40960 │ Yes │ │
│ │ 4 │ llama3.2:latest │ 1925.8 │ llama │ 131072 │ Yes │ │
│ │ 5 │ mistral-small3.2:latest │ 14474.3 │ mistral3 │ 131072 │ Yes │ │
│ ╰─────┴───────────────────────────┴──────────────┴─────────────────┴──────────┴───────╯ │
│ │
│ 💡 Options: │
│ • 🔢 Select model (1-5) │
│ • 👋 Type 'q' to quit ││ │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice: 3
╭─ Success ───────────────────────────────────────────────────────────────────────────────────────────────╮
│ Summary model set to 'qwen3:latest' for this session │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─ 💬 Chat Session ───────────────────────────────────────────────────────────────────────────────────────╮
│ Session ID: a102ba9d9a │
│ Model: gpt-oss:20b │
│ Summary Model: qwen3:latest │
│ Context Window: No valid context data in session │
│ Tools: None │
│ Tool Policy: Not applicable │
│ Markdown: Enabled │
│ Thinking Blocks: Enabled │
│ Summary: Not available │
│ │
│ 💡 Available Commands: │
│ • /menu - Open the main menu │
│ • /status - Show session information │
│ • /edit - Edit a previous message │
│ • /exit or /quit - Exit the application │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭────────╮
│ 🧑 You │
╰────────╯
Chat Session Menu
During a chat session it is possible to open the chat menu with the command /menu. Select one of the options by typing the number (e.g. 1).
Tip:
Each of the available options in this menu is also a command itself and can therefore be directly executed without opening the chat menu. For example, typing
/1will switch to a different chat session. Or/5to toggle the tool feature and open the tool menu.
╭────────╮
│ 🧑 You │
╰────────╯
/menu
╭─ ⌨️ Available Commands ───────────────────────────────────────────────────────────────────────╮
│ ╭──────────────┬────────────────────────┬─────────────────────────────────────╮ │
│ │ /1 │ 💬 Switch Sessions │ Change to different chat session │ │
│ │ /2 │ 🤖 Change Model │ Select a different AI model │ │
│ │ /3 │ 📝 Toggle Markdown │ Enable/disable markdown rendering │ │
│ │ /4 │ 🤔 Toggle Thinking │ Show/hide thinking blocks │ │
│ │ /5 │ 📂 Enable Tools │ Select tools to use │ │
│ │ /6 │ 🔧 System Prompt │ Select different system prompt │ │
│ │ /help │ 📚 Help │ Show all available commands │ │
│ │ /quit │ 👋 Exit │ Exit the application │ │
│ ╰──────────────┴────────────────────────┴─────────────────────────────────────╯ │
╰───────────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice:
System Prompts
System prompts have to be provided within the /system_prompts folder at the terminals root directory. Only *.md or *.txt files are supported.
Find more here: User Guide System Prompts
The system prompts can be selected when creating a new chat session:
╭─ 🔧 System Prompts ──────────────────────────────────────────────────────────────────────╮
│ ╭─────┬─────────────────┬─────────────────────────────────────┬────────────╮ │
│ │ # │ Filename │ Preview │ Word Count │ │
│ ├─────┼─────────────────┼─────────────────────────────────────┼────────────┤ │
│ │ 1 │ AGENTS.md │ You are a helpful agent │ 5 │ │
│ ╰─────┴─────────────────┴─────────────────────────────────────┴────────────╯ │
│ │
│ 💡 Options: │
│ • 📝 Select system prompt (1-1) │
│ • 🆕 Type 'no' for no system prompt │
│ • 🗑️ Type '/delete <number>' to delete a system prompt │
│ • 👋 Type 'q' to quit │
╰──────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice:
... or by opening the chat session menu and selecting the corresponding option
╭────────╮
│ 🧑 You │
╰────────╯
/menu
╭─ ⌨️ Available Commands ──────────────────────────────────────────────────────────────────╮
│ ╭──────────────┬────────────────────────┬─────────────────────────────────────╮ │
│ │ /1 │ 💬 Switch Sessions │ Change to different chat session │ │
│ │ /2 │ 🤖 Change Model │ Select a different AI model │ │
│ │ /3 │ 📝 Toggle Markdown │ Enable/disable markdown rendering │ │
│ │ /4 │ 🤔 Toggle Thinking │ Show/hide thinking blocks │ │
│ │ /5 │ 📂 Enable Tools │ Select tools to use │ │
│ │ /6 │ 🔧 System Prompt │ Select different system prompt │ │
│ │ /quit /q │ 👋 Exit │ Exit the menu │ │
│ ╰──────────────┴────────────────────────┴─────────────────────────────────────╯ │
╰──────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice: 6
╭─ 🔧 System Prompts ──────────────────────────────────────────────────────────────────────╮
│ ╭─────┬─────────────────┬─────────────────────────────────────┬────────────╮ │
│ │ # │ Filename │ Preview │ Word Count │ │
│ ├─────┼─────────────────┼─────────────────────────────────────┼────────────┤ │
│ │ 1 │ AGENTS.md │ You are a helpful agent │ 5 │ │
│ ╰─────┴─────────────────┴─────────────────────────────────────┴────────────╯ │
│ │
│ 💡 Options: │
│ • 📝 Select system prompt (1-1) │
│ • 🆕 Type 'no' for no system prompt │
│ • 🗑️ Type '/delete <number>' to delete a system prompt │
│ • 👋 Type 'q' to quit │
╰──────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice:
... or by typing the shortcut /system into the chat
╭────────╮
│ 🧑 You │
╰────────╯
/system
╭─ 🔧 System Prompts ──────────────────────────────────────────────────────────────────────╮
│ ╭─────┬─────────────────┬─────────────────────────────────────┬────────────╮ │
│ │ # │ Filename │ Preview │ Word Count │ │
│ ├─────┼─────────────────┼─────────────────────────────────────┼────────────┤ │
│ │ 1 │ AGENTS.md │ You are a helpful agent │ 5 │ │
│ ╰─────┴─────────────────┴─────────────────────────────────────┴────────────╯ │
│ │
│ 💡 Options: │
│ • 📝 Select system prompt (1-1) │
│ • 🆕 Type 'no' for no system prompt │
│ • 🗑️ Type '/delete <number>' to delete a system prompt │
│ • 👋 Type 'q' to quit │
╰──────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice:
Edit Menu
With the in-chat command /edit you can edit messages from the current chat history.
╭────────╮
│ 🧑 You │
╰────────╯
/edit
✏️ Edit Message
╭─ ✏️ Edit Messages ────────────────────────────────────────────────────────────────────────────────────────╮
│ ╭─────┬──────────────┬───────────────────────────────────────────────────────────────────────────╮ │
│ │ # │ Role │ Preview │ │
│ ├─────┼──────────────┼───────────────────────────────────────────────────────────────────────────┤ │
│ │ 1 │ 🧑 User │ Who was the first Avenger in the MCU? │ │
│ │ - │ 🤖 Assistant │ **Captain America (Steve Rogers)** is widely considered the first Aven... │ │
│ ╰─────┴──────────────┴───────────────────────────────────────────────────────────────────────────╯ │
╰────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ Select a user message (1-1) or 'q' to cancel │
╰────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
Disclaimer
Editing a message will continue the chat from this point. All previously trailing messages - whether user or assistant - will be deleted.
Tool Menu
The tool feature depends on the existence of a /tools folder in the terminal's root directory and python functions. Tools are not enabled by default.
Get more information how to use the tool feature here: User Guide: Tools
╭────────╮
│ 🧑 You │
╰────────╯
/5
No tools currently selected
╭─ 🛠️ Tool Selection ──────────────────────────────────────────────────────────────────────╮
│ Individual Tools │
│ ╭───────┬───────────────────────────┬──────────────────────────────────────────────────╮ │
│ │ # │ Tool Name │ Description │ │
│ ├───────┼───────────────────────────┼──────────────────────────────────────────────────┤ │
│ │ 1 │ add_numbers │ Add two numbers together. │ │
│ │ 2 │ subtract_numbers │ Subtract the second number from the first │ │
│ │ │ │ number. │ │
│ │ 3 │ multiply_numbers │ Multiply two numbers together. │ │
│ │ 4 │ divide_numbers │ Divide the first number by the second number. │ │
│ │ 5 │ power_calculation │ Calculate base raised to the power of exponent. │ │
│ │ 6 │ square_root │ Calculate the square root of a number. │ │
│ │ 7 │ calculate_percentage │ Calculate what percentage one number is of │ │
│ │ │ │ another. │ │
│ │ 8 │ get_current_time │ Get the current date and time. │ │
│ │ 9 │ generate_random_number │ Generate a random number between two values. │ │
│ │ 10 │ flip_coin │ Flip a virtual coin. │ │
│ │ 11 │ count_words │ Count the number of words in a text. │ │
│ │ 12 │ reverse_text │ Reverse the given text. │ │
│ │ 13 │ roll_dice │ Roll one or more dice with specified number of │ │
│ │ │ │ sides. │ │
│ ╰───────┴───────────────────────────┴──────────────────────────────────────────────────╯ │
│ │
│ Tool Groups │
│ ╭──────────┬───────────────────────────┬───────────────────────────────────────────────╮ │
│ │ Letter │ Group Name │ Tools Included │ │
│ ├──────────┼───────────────────────────┼───────────────────────────────────────────────┤ │
│ │ a │ basic_calculator │ add_numbers, subtract_numbers, │ │
│ │ │ │ multiply_numbers... │ │
│ │ b │ fun_tools │ flip_coin, roll_dice, reverse_text, │ │
│ │ │ │ generate_ra... │ │
│ │ c │ math │ add_numbers, subtract_numbers, │ │
│ │ │ │ multiply_numbers... │ │
│ │ d │ utilities │ get_current_time, generate_random_number, │ │
│ │ │ │ flip_... │ │
│ ╰──────────┴───────────────────────────┴───────────────────────────────────────────────╯ │
│ │
│ 💡 Options: │
│ • 🔢 Select tools by numbers (e.g., 1,3,4 or 1-3) │
│ • 📂 Select a group by letter (e.g., a) │
│ • ❌ Type 'none' to clear selection │
│ • 🔄 Type 'reload' to refresh tools │
│ • ↩️ Press Enter to keep current selection │
│ • 👋 Type 'q' to cancel │
╰──────────────────────────────────────────────────────────────────────────────────────────╯
Enter your choice:
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
mochi_coco-0.6.1.tar.gz
(194.1 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
mochi_coco-0.6.1-py3-none-any.whl
(108.0 kB
view details)
File details
Details for the file mochi_coco-0.6.1.tar.gz.
File metadata
- Download URL: mochi_coco-0.6.1.tar.gz
- Upload date:
- Size: 194.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
uv/0.7.20
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f68b30bb125b35be6e6509636aac53e25eef4735cc2f33f5860f6440adbd09cb
|
|
| MD5 |
3a8baf6105d26a01f035e06ae2b05505
|
|
| BLAKE2b-256 |
344aff04d105b33e80bcd18ed3a02726cbd8a11ba2d39e26314716e8ad8fd081
|
File details
Details for the file mochi_coco-0.6.1-py3-none-any.whl.
File metadata
- Download URL: mochi_coco-0.6.1-py3-none-any.whl
- Upload date:
- Size: 108.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
uv/0.7.20
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
a55c686809d8ac0e92019ff34572beeb081b6f2d40b26dd7e9b04483209635af
|
|
| MD5 |
1708983793436c48e5c6816d4a199637
|
|
| BLAKE2b-256 |
a9211b24e204fc185df9f13d747c884f16681bb17bd412d6ecc6fd9cdf35cd42
|
