janito 3.12.2__tar.gz → 3.13.0__tar.gz

{janito-3.12.2 → janito-3.13.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: janito
-Version: 3.12.2
+Version: 3.13.0
 Summary: A new Python package called janito.
 Author-email: João Pinto <janito@ikignosis.org>
 Project-URL: Homepage, https://github.com/ikignosis/janito

{janito-3.12.2 → janito-3.13.0}/docs/CHANGELOG.md

@@ -1,49 +1,49 @@
-# Changelog
-
-All notable changes to Janito will be documented in this file.
-
-The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
-and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
-
-## [Unreleased]
-
-### Removed
-
-- **Breaking Change**: Removed `--role` argument and interactive profile selection
-  - The `--role` argument has been completely removed from the CLI
-  - Interactive profile selection has been removed from chat mode
-  - Use `--profile <name>` or shorthand flags like `--developer` and `--market` instead
-  - Default behavior now uses the Developer profile when no profile is specified
-
-### Changed
-
-- Updated documentation to reflect removal of role argument
-- Added comprehensive profile documentation in `PROFILES.md`
-- Simplified profile selection to use explicit flags only
-
-## [Previous Versions]
-
-### Added
-
-- Initial support for profiles and roles
-- Interactive profile selection in chat mode
-- `--role` argument for specifying developer roles
-- `--profile` argument for system prompt templates
-- `--developer` and `--market` shorthand flags
-
-### Available Profiles
-
-- **Developer**: Optimized for software development tasks
-- **Market Analyst**: Specialized for market analysis and business insights
-
-### Supported Providers
-
-- Moonshot AI (default)
-- OpenAI
-- Anthropic
-- IBM WatsonX
-- Google AI
-
----
-
+# Changelog
+
+All notable changes to Janito will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [3.12.3] - 2025-09-12
+
+### Removed
+
+- **Breaking Change**: Removed `--role` argument and interactive profile selection
+  - The `--role` argument has been completely removed from the CLI
+  - Interactive profile selection has been removed from chat mode
+  - Use `--profile <name>` or shorthand flags like `--developer` and `--market` instead
+  - Default behavior now uses the Developer profile when no profile is specified
+
+### Changed
+
+- Updated documentation to reflect removal of role argument
+- Added comprehensive profile documentation in `PROFILES.md`
+- Simplified profile selection to use explicit flags only
+
+## [Previous Versions]
+
+### Added
+
+- Initial support for profiles and roles
+- Interactive profile selection in chat mode
+- `--role` argument for specifying developer roles
+- `--profile` argument for system prompt templates
+- `--developer` and `--market` shorthand flags
+
+### Available Profiles
+
+- **Developer**: Optimized for software development tasks
+- **Market Analyst**: Specialized for market analysis and business insights
+
+### Supported Providers
+
+- Moonshot AI (default)
+- OpenAI
+- Anthropic
+- IBM WatsonX
+- Google AI
+
+---
+
 For detailed information about profiles and their usage, see [PROFILES.md](PROFILES.md).

janito-3.13.0/docs/README.md

@@ -0,0 +1,41 @@
+# Janito LLM Platform Documentation
+
+Welcome to the official documentation for Janito, a flexible and extensible platform for interacting with large language models (LLMs) from multiple providers.
+
+## Table of Contents
+
+- [Supported Models](models/supported_models.md)
+- [Providers](providers.md)
+- [CLI Commands](cli.md)
+- [Configuration](configuration.md)
+- [Tools & Plugins](tools.md)
+
+## Overview
+
+Janito provides a unified interface to access state-of-the-art LLMs from providers such as OpenAI, Anthropic, Google Gemini, Mistral, Alibaba Qwen, Z.AI, DeepSeek, Moonshot, IBM WatsonX, and Azure OpenAI.
+
+All model specifications are automatically synchronized with the codebase. To see the latest supported models, visit the [Supported Models](models/supported_models.md) page.
+
+## Getting Started
+
+1. Install Janito:
+   ```bash
+   pip install janito
+   ```
+
+2. Set your API key:
+   ```bash
+   janito set-api-key --provider openai --key YOUR_API_KEY
+   ```
+
+3. List available models:
+   ```bash
+   janito list-models --provider openai
+   ```
+
+4. Start interactive chat:
+   ```bash
+   janito chat
+   ```
+
+For detailed usage, refer to the linked guides above.

janito-3.13.0/docs/anthropic-setup.md

@@ -0,0 +1,63 @@
+# Anthropic Setup
+
+To use Anthropic models with Janito, follow these steps:
+
+## 1. Get Your API Key
+
+1. Sign up at [https://console.anthropic.com/](https://console.anthropic.com/)
+2. Navigate to your API keys page
+3. Create a new API key or copy an existing one
+
+## 2. Set the API Key
+
+Set your Anthropic API key using the CLI:
+
+```bash
+janito set-api-key --provider anthropic --key YOUR_API_KEY_HERE
+```
+
+Or set it as an environment variable:
+
+```bash
+export ANTHROPIC_API_KEY="YOUR_API_KEY_HERE"
+```
+
+## 3. Available Models
+
+Janito supports the following Anthropic models:
+
+- `claude-opus-4-20250514` - Most capable model for complex tasks
+- `claude-sonnet-4-20250514` - Balanced performance and cost
+- `claude-3-7-sonnet-20250219` - High-performance model
+- `claude-3-5-haiku-20241022` - Fast and cost-effective
+- `claude-3-5-sonnet-20241022` - Balanced speed and capability
+- `claude-3-haiku-20240307` - Fastest model for simple tasks
+
+## 4. Usage Examples
+
+### List available models
+
+```bash
+janito list-models --provider anthropic
+```
+
+### Use a specific model
+
+```bash
+janito chat --provider anthropic --model claude-3-5-sonnet-20241022
+```
+
+### Set as default provider
+
+```bash
+janito set-config provider=anthropic
+```
+
+## Notes
+
+- Anthropic is accessed through an OpenAI-compatible API endpoint
+- The default model is `claude-3-7-sonnet-20250219`
+- All models support streaming responses
+- Context window sizes vary by model (see [Supported Models](models/supported_models.md) for details)
+
+For more information, visit the [Anthropic documentation](https://docs.anthropic.com/).

janito-3.13.0/docs/cli.md

@@ -0,0 +1,129 @@
+# CLI Commands
+
+Janito provides a comprehensive command-line interface (CLI) for managing and interacting with LLM providers.
+
+## Core Commands
+
+### `list-models`
+
+List all supported models for a given provider.
+
+```bash
+# List models for OpenAI
+janito list-models --provider openai
+
+# List models for the default provider
+janito list-models
+```
+
+### `list-providers`
+
+List all available providers and their status.
+
+```bash
+janito list-providers
+```
+
+### `set-api-key`
+
+Set your API key for a specific provider.
+
+```bash
+janito set-api-key --provider openai --key sk-your-key-here
+```
+
+### `chat`
+
+Start an interactive chat session with the default model.
+
+```bash
+janito chat
+```
+
+### `show-config`
+
+Display current configuration settings.
+
+```bash
+janito show-config
+```
+
+### `ping-providers`
+
+Test connectivity to all providers.
+
+```bash
+janito ping-providers
+```
+
+## Provider-Specific Commands
+
+### `list-providers-region`
+
+List providers available in a specific region.
+
+```bash
+janito list-providers-region --region us-east-1
+```
+
+### `list-tools`
+
+List all available tools and plugins.
+
+```bash
+janito list-tools
+```
+
+## Configuration Commands
+
+### `set-config`
+
+Set a configuration value.
+
+```bash
+janito set-config provider=openai
+janito set-config azure_deployment_name=my-deployment
+```
+
+### `show-config`
+
+Show current configuration.
+
+```bash
+janito show-config
+```
+
+### `unset-config`
+
+Remove a configuration value.
+
+```bash
+janito unset-config provider
+```
+
+## Advanced Commands
+
+### `model-selection`
+
+Interactively select a model from available options.
+
+```bash
+janito model-selection
+```
+
+### `enable-plugin` / `disable-plugin`
+
+Enable or disable a plugin.
+
+```bash
+janito enable-plugin webtools
+janito disable-plugin filemanager
+```
+
+For detailed help on any command, use:
+
+```bash
+janito COMMAND --help
+```
+
+> **Note**: The `list-models` command automatically reflects the latest model specifications defined in the codebase. No manual documentation update is required when new models are added.

janito-3.13.0/docs/configuration.md

@@ -0,0 +1,104 @@
+# Configuration
+
+Janito's behavior is controlled through a combination of environment variables, configuration files, and CLI commands.
+
+## Configuration Sources (Priority Order)
+
+1. **Command-line arguments** (highest priority)
+2. **Environment variables**
+3. **Configuration file** (`~/.janito/config.yaml`)
+4. **Default values** (lowest priority)
+
+## Environment Variables
+
+| Variable | Description | Example |
+|----------|-------------|---------|
+| `OPENAI_API_KEY` | OpenAI API key | `sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `ANTHROPIC_API_KEY` | Anthropic API key | `sk-ant-api03-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `GOOGLE_API_KEY` | Google Gemini API key | `AIzaSyBxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `MISTRAL_API_KEY` | Mistral API key | `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `CEREBRAS_API_KEY` | Cerebras API key | `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `ZAI_API_KEY` | Z.AI API key | `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `ALIBABA_API_KEY` | Alibaba Qwen API key | `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `DEEPSEEK_API_KEY` | DeepSeek API key | `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `MOONSHOT_API_KEY` | Moonshot API key | `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `WATSONX_API_KEY` | IBM WatsonX API key | `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `WATSONX_PROJECT_ID` | IBM WatsonX project ID | `xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx` |
+| `WATSONX_SPACE_ID` | IBM WatsonX space ID | `xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx` |
+| `AZURE_OPENAI_API_KEY` | Azure OpenAI API key | `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx` |
+| `JANITO_PROVIDER` | Default provider | `openai` |
+
+## Configuration File
+
+The configuration file is located at `~/.janito/config.yaml` and has the following structure:
+
+```yaml
+provider: openai
+azure_deployment_name: my-deployment
+max_tokens: 4096
+default_temperature: 0.7
+enable_tools: true
+```
+
+You can modify this file directly or use the CLI:
+
+```bash
+# Set default provider
+janito set-config provider=openai
+
+# Set Azure deployment name
+janito set-config azure_deployment_name=my-deployment
+
+# View current config
+janito show-config
+```
+
+## Provider-Specific Configuration
+
+### Azure OpenAI
+
+For Azure OpenAI, you must specify your deployment name:
+
+```bash
+janito set-config azure_deployment_name=my-gpt4o-deployment
+```
+
+This maps to the `--model` parameter in CLI commands:
+
+```bash
+janito chat --provider azure_openai
+# Internally uses model: my-gpt4o-deployment
+```
+
+### IBM WatsonX
+
+For IBM WatsonX, you need both API key and project ID:
+
+```bash
+janito set-config watsonx_project_id=your-project-id
+janito set-config watsonx_space_id=your-space-id
+```
+
+## Model Selection
+
+When no model is specified, Janito uses the provider's default model:
+
+- **OpenAI**: `gpt-5`
+- **Anthropic**: `claude-3-7-sonnet-20250219`
+- **Google**: `gemini-2.5-flash`
+- **Mistral**: `mistral-large-latest`
+- **Cerebras**: `qwen-3-coder-480b`
+- **Z.AI**: `glm-4.5`
+- **Alibaba**: `qwen3-next-80b-a3b-instruct`
+- **DeepSeek**: `deepseek-chat`
+- **Moonshot**: `kimi-k2-0905-preview`
+- **IBM WatsonX**: `ibm/granite-3-3-8b-instruct`
+- **Azure OpenAI**: `azure_openai_deployment`
+
+You can override the default model using `--model MODEL_NAME` on any command:
+
+```bash
+janito chat --provider openai --model gpt-4o
+```
+
+> **Note**: The list of available models is automatically synchronized with the codebase. Use `janito list-models --provider PROVIDER` to see all available options.

janito-3.13.0/docs/google-setup.md

@@ -0,0 +1,62 @@
+# Google Gemini Setup
+
+To use Google Gemini models with Janito, follow these steps:
+
+## 1. Get Your API Key
+
+1. Go to the [Google Cloud Console](https://console.cloud.google.com/)
+2. Create a new project or select an existing one
+3. Enable the Generative Language API
+4. Navigate to "APIs & Services" → "Credentials"
+5. Create an API key
+
+## 2. Set the API Key
+
+Set your Google API key using the CLI:
+
+```bash
+janito set-api-key --provider google --key YOUR_API_KEY_HERE
+```
+
+Or set it as an environment variable:
+
+```bash
+export GOOGLE_API_KEY="YOUR_API_KEY_HERE"
+```
+
+## 3. Available Models
+
+Janito supports the following Google Gemini models:
+
+- `gemini-2.5-flash` - Fastest model for simple tasks
+- `gemini-2.5-pro` - Most capable model for complex tasks
+- `gemini-2.5-flash-lite-preview-06-17` - Lightweight version for low-latency applications
+
+## 4. Usage Examples
+
+### List available models
+
+```bash
+janito list-models --provider google
+```
+
+### Use a specific model
+
+```bash
+janito chat --provider google --model gemini-2.5-pro
+```
+
+### Set as default provider
+
+```bash
+janito set-config provider=google
+```
+
+## Notes
+
+- Google Gemini is accessed through an OpenAI-compatible API endpoint
+- The default model is `gemini-2.5-flash`
+- All models support streaming responses and tool usage
+- Context window sizes vary by model (see [Supported Models](models/supported_models.md) for details)
+
+For more information, visit the [Google AI documentation](https://ai.google.dev/).

janito-3.13.0/docs/index.md

@@ -0,0 +1,47 @@
+# Welcome to Janito
+
+Janito is a flexible and extensible platform for interacting with large language models (LLMs) from multiple providers.
+
+## Key Features
+
+- **Multi-provider support**: Access models from OpenAI, Anthropic, Google Gemini, Mistral, Alibaba Qwen, Z.AI, DeepSeek, Moonshot, IBM WatsonX, and Azure OpenAI
+- **Unified interface**: Consistent CLI and API across all providers
+- **Tool integration**: Built-in tools for file management, web access, code execution, and more
+- **Plugin system**: Extend functionality with modular plugins
+- **Security first**: Permission controls and sandboxed operations
+- **Automatic documentation**: Supported models are automatically synchronized with the codebase
+
+## Getting Started
+
+1. Install Janito:
+   ```bash
+   pip install janito
+   ```
+
+2. Set your API key:
+   ```bash
+   janito set-api-key --provider openai --key YOUR_API_KEY
+   ```
+
+3. List available models:
+   ```bash
+   janito list-models --provider openai
+   ```
+
+4. Start interactive chat:
+   ```bash
+   janito chat
+   ```
+
+## Documentation
+
+Explore our comprehensive documentation:
+
+- [Installation & Setup](guides/installation.md)
+- [Configuration](configuration.md)
+- [CLI Commands](cli.md)
+- [Supported Models](models/supported_models.md)
+- [Tools & Plugins](tools.md)
+- [Developer Guide](guides/developing.md)
+
+> **Note**: The list of supported models is automatically generated from the codebase. When new models are added, this documentation updates automatically.*

janito-3.13.0/docs/models/supported_models.md

@@ -0,0 +1,134 @@
+# Supported LLM Models
+
+This document lists all supported large language models (LLMs) across providers, as defined in the codebase.
+
+## OpenAI
+
+| Model Name | Context Window | Max Response | Thinking Supported |
+|------------|----------------|--------------|--------------------|
+| `gpt-3.5-turbo` | 16,385 | 4,096 | No |
+| `gpt-4.1` | 1,047,576 | 32,768 | No |
+| `gpt-4.1-mini` | 1,047,576 | 32,768 | No |
+| `gpt-4.1-nano` | 1,047,576 | 32,768 | No |
+| `gpt-4-turbo` | 128,000 | N/A | No |
+| `gpt-4o` | 128,000 | 4,096 | No |
+| `gpt-4o-mini` | 128,000 | 16,384 | No |
+| `o3-mini` | 200,000 | 100,000 | Yes |
+| `o3` | 200,000 | 100,000 | Yes |
+| `o4-mini` | 200,000 | 100,000 | Yes |
+| `gpt-5` | 200,000 | 100,000 | Yes |
+| `gpt-5-mini` | 200,000 | 100,000 | Yes |
+| `gpt-5-nano` | 200,000 | 100,000 | Yes |
+
+> **Note**: `gpt-4-turbo-alt` is a duplicate entry for internal use and should not be selected by users.
+
+## Anthropic
+
+| Model Name | Max Response | Thinking Supported |
+|------------|--------------|--------------------|
+| `claude-opus-4-20250514` | 32,000 | No |
+| `claude-sonnet-4-20250514` | 64,000 | No |
+| `claude-3-7-sonnet-20250219` | 64,000 | No |
+| `claude-3-5-haiku-20241022` | 8,192 | No |
+| `claude-3-5-sonnet-20241022` | 8,192 | No |
+| `claude-3-haiku-20240307` | 4,096 | No |
+
+## Google Gemini
+
+| Model Name | Max Response | Max COT | Thinking Supported |
+|------------|--------------|---------|--------------------|
+| `gemini-2.5-flash` | 8,192 | 24,576 | Yes |
+| `gemini-2.5-pro` | 65,536 | 196,608 | Yes |
+| `gemini-2.5-flash-lite-preview-06-17` | 64,000 | 192,000 | Yes |
+
+## Mistral
+
+| Model Name | Context Window | Max Input | Max Response | Thinking Supported |
+|------------|----------------|-----------|--------------|--------------------|
+| `codestral-latest` | 256,000 | 250,000 | 4,096 | No |
+| `codestral-2405` | 256,000 | 250,000 | 4,096 | No |
+| `mistral-small-latest` | 32,000 | 28,000 | 4,096 | No |
+| `mistral-medium-latest` | 32,000 | 28,000 | 4,096 | No |
+| `mistral-large-latest` | 128,000 | 120,000 | 4,096 | No |
+| `devstral-small-latest` | 128,000 | 120,000 | 4,096 | No |
+| `devstral-medium-latest` | 128,000 | 120,000 | 4,096 | No |
+
+## Cerebras
+
+| Model Name | Max Input | Max Response | Description |
+|------------|-----------|--------------|-------------|
+| `qwen-3-32b` | 128,000 | 16,384 | Qwen 3 32B model for general instruction following |
+| `qwen-3-235b-a22b-instruct-2507` | 128,000 | 16,384 | Qwen 3 235B A22B instruction-tuned model (preview) |
+| `qwen-3-235b-a22b-thinking-2507` | 128,000 | 16,384 | Qwen 3 235B A22B thinking model for reasoning tasks (preview) |
+| `qwen-3-coder-480b` | 128,000 | 16,384 | Qwen 3 Coder 480B model for programming tasks (preview) |
+| `gpt-oss-120b` | 128,000 | 16,384 | GPT-OSS 120B open-source model (preview) |
+
+## Z.AI
+
+| Model Name | Context Window | Max Input | Max COT | Max Response | Thinking Supported | Supports Tools | Supports Images |
+|------------|----------------|-----------|---------|--------------|--------------------|----------------|-----------------|
+| `glm-4.5` | 128,000 | 128,000 | 4,096 | 4,096 | Yes | Yes | Yes |
+| `glm-4.5-air` | 128,000 | 128,000 | 4,096 | 4,096 | Yes | Yes | Yes |
+
+## Alibaba Qwen
+
+| Model Name | Context Window | Max Response | Category |
+|------------|----------------|--------------|----------|
+| `qwen-turbo` | 1,008,192 | 8,192 | Alibaba Qwen Turbo Model (OpenAI-compatible) |
+| `qwen-plus` | 131,072 | 8,192 | Alibaba Qwen Plus Model (OpenAI-compatible) |
+| `qwen-flash` | 1,000,000 | 8,192 | Alibaba Qwen Flash Model (OpenAI-compatible) |
+| `qwen-max` | 32,768 | 8,192 | Alibaba Qwen Max Model (OpenAI-compatible) |
+| `qwen3-coder-plus` | 1,048,576 | 65,536 | Alibaba Qwen3 Coder Plus Model (OpenAI-compatible) |
+| `qwen3-coder-480b-a35b-instruct` | 262,144 | 65,536 | Alibaba Qwen3 Coder 480B A35B Instruct Model (OpenAI-compatible) |
+| `qwen3-235b-a22b-thinking-2507` | 131,072 | 32,768 | Alibaba Qwen3 235B A22B Thinking Model (OpenAI-compatible) |
+| `qwen3-235b-a22b-instruct-2507` | 129,024 | 32,768 | Alibaba Qwen3 235B A22B Instruct Model (OpenAI-compatible) |
+| `qwen3-30b-a3b-thinking-2507` | 126,976 | 32,768 | Alibaba Qwen3 30B A3B Thinking Model (OpenAI-compatible) |
+| `qwen3-30b-a3b-instruct-2507` | 129,024 | 32,768 | Alibaba Qwen3 30B A3B Instruct Model (OpenAI-compatible) |
+| `qwen3-next-80b-a3b-instruct` | 262,144 | 65,536 | Alibaba Qwen3-Max Preview (256K) - 80B A3B Instruct Model (OpenAI-compatible) |
+| `qwen3-next-80b-a3b-thinking` | 262,144 | 65,536 | Alibaba Qwen3-Max Preview (256K) - 80B A3B Thinking Model (OpenAI-compatible) |
+| `qwen3-max-preview` | 262,144 | 65,536 | Alibaba Qwen3-Max Preview (256K) - Standard Model (OpenAI-compatible) |
+
+## DeepSeek
+
+| Model Name | Context Window | Max Tokens | Description |
+|------------|----------------|------------|-------------|
+| `deepseek-chat` | 8,192 | 4,096 | DeepSeek Chat Model (OpenAI-compatible) |
+| `deepseek-reasoner` | 8,192 | 4,096 | DeepSeek Reasoner Model (OpenAI-compatible) |
+| `deepseek-v3.1` | 131,072 | 4,096 | DeepSeek V3.1 Model (128K context, OpenAI-compatible) |
+| `deepseek-v3.1-base` | 131,072 | 4,096 | DeepSeek V3.1 Base Model (128K context, OpenAI-compatible) |
+| `deepseek-r1` | 131,072 | 4,096 | DeepSeek R1 Model (128K context, OpenAI-compatible) |
+
+## Moonshot
+
+| Model Name | Context Window | Max Input | Max Response |
+|------------|----------------|-----------|--------------|
+| `kimi-k2-0711-preview` | 128,000 | 100,000 | 4,096 |
+| `kimi-k2-turbo-preview` | 128,000 | 100,000 | 4,096 |
+| `kimi-k2-0905-preview` | 128,000 | 100,000 | 4,096 |
+
+## IBM WatsonX
+
+| Model Name | Context Window | Max Input | Max Response | Max COT | Thinking Supported |
+|------------|----------------|-----------|--------------|---------|--------------------|
+| `openai/gpt-oss-120b` | 128,000 | 128,000 | 4,096 | 4,096 | Yes |
+| `ibm/granite-3-8b-instruct` | 128,000 | 128,000 | 4,096 | 4,096 | No |
+| `ibm/granite-3-3-8b-instruct` | 128,000 | 128,000 | 4,096 | 4,096 | No |
+| `meta-llama/llama-3-1-70b-instruct` | 128,000 | 128,000 | 4,096 | 4,096 | No |
+| `meta-llama/llama-3-3-70b-instruct` | 128,000 | 128,000 | 4,096 | 4,096 | No |
+| `mistralai/mistral-large` | 128,000 | 128,000 | 4,096 | 4,096 | No |
+| `mistralai/mistral-large-2407` | 128,000 | 128,000 | 4,096 | 4,096 | No |
+| `openai/gpt-oss-20b` | 128,000 | 128,000 | 4,096 | 4,096 | Yes |
+
+## Azure OpenAI
+
+Azure OpenAI supports any deployment name configured by the user. The following are known model mappings:
+
+| Model Name | Mapped To | Context Window | Max Response |
+|------------|-----------|----------------|--------------|
+| `azure_openai_deployment` | gpt-4o | 128,000 | 4,096 |
+
+> **Note**: Azure OpenAI deployments are user-defined. Use `--provider azure_openai --model YOUR_DEPLOYMENT_NAME` to use any valid deployment.
+
+---
+
+*Last updated from source code: September 2025*