> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zerotwo.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Choose a Model
> Browse 95+ AI models in ZeroTwo, understand the categories and quotas, and pick the right model for your task.
ZeroTwo gives you access to **95+ models from 19 providers** — all accessible from a single model picker inside the chat interface. This page explains how to open the picker, what the model categories mean, how quotas work, and how to choose the right model for your task.
***
## How to Open the Model Picker
The name of the currently active model appears as a clickable label in the **prompt bar** at the bottom of the chat area. Click it to open the model picker overlay.
Inside the picker you can:
* Browse models grouped by **Standard**, **Premium**, and **Reasoning** categories
* Filter by **provider** (Anthropic, OpenAI, Google, Mistral, xAI, and more)
* Search by model name using the search field
* **Star** a model to pin it to the top of the list for quick access
* Click any model name to select it immediately
***
## Model Categories
| Category | Examples | Best for |
| ------------- | ----------------------------------------------------------------- | ------------------------------------------------------------------- |
| **Standard** | GPT-4o-mini, Gemini 2.5 Flash Lite, Mistral Small, Grok 4 Fast | Everyday tasks, fast responses, drafting, Q\&A |
| **Premium** | Claude Sonnet 4.6, GPT-5, Gemini 2.5 Pro, Grok-4, Claude Opus 4.6 | Complex reasoning, long documents, advanced coding, nuanced writing |
| **Reasoning** | o3, o4-mini (OpenAI), Claude with extended thinking | Math, logic, step-by-step problem solving, algorithm design |
Standard models are fast and cost-efficient. They do not count against your monthly premium message quota and are available on all plans — including Free (subject to the 20 messages/day overall daily limit).
Standard models are a great choice for:
* Quick questions and answers
* Drafting emails, summaries, and social posts
* Iterating rapidly through ideas
* Tasks where response speed matters more than depth
When your premium quota is exhausted, ZeroTwo automatically falls back to a standard model. See [Fallback Models](#fallback-models) below.
Premium models are the most capable models from each provider. Each message sent with a premium model counts against your monthly quota.
| Plan | Premium Messages / Month |
| ---------- | ----------------------------------- |
| Free | Not included (standard models only) |
| Pro | 555 premium messages |
| Pro 2x | 1,150 premium messages |
| Plus Ultra | Unlimited |
| Business | Unlimited |
Premium models excel at:
* Complex multi-step reasoning
* Long-document analysis and summarization
* High-quality creative writing
* Advanced code generation and review
* Tasks requiring nuanced judgment or deep knowledge
Notable premium models available in ZeroTwo:
* **Anthropic:** Claude Sonnet 4.6, Claude Opus 4.6
* **OpenAI:** GPT-4.1, GPT-5
* **Google:** Gemini 2.5 Pro, Gemini 3.1
* **xAI:** Grok-3, Grok-4
* **DeepSeek:** DeepSeek latest
Reasoning models are specialized for tasks requiring careful multi-step logic. They deliberate before producing an answer — which makes them slower but significantly more accurate on hard problems.
Use reasoning models for:
* Mathematics and proofs
* Algorithm design and analysis
* Logic puzzles and deductive reasoning
* Complex debugging where root cause is non-obvious
* Scientific problem-solving
Available reasoning models:
* **o3, o4-mini** (OpenAI)
* **Claude models with extended thinking enabled** (Anthropic)
* **DeepSeek R-series**
Reasoning models take longer to respond than standard or premium models. For creative writing, quick drafts, or conversational tasks, a premium model is usually a better choice.
***
## Premium Message Quotas
Any message sent using a **rate-limited premium model** counts as one premium message against your monthly quota. The quota resets on your billing cycle date each month.
| Plan | Monthly Premium Messages |
| -------------- | ------------------------ |
| **Free** | 0 (standard models only) |
| **Pro** | 555 |
| **Pro 2x** | 1,150 |
| **Plus Ultra** | Unlimited |
| **Business** | Unlimited |
You can check your remaining quota at any time in **Settings → Account**.
***
## Fallback Models
When your monthly premium quota is exhausted, ZeroTwo automatically falls back to a standard model for the remainder of the billing cycle. A notice appears in the chat when this happens.
Current fallback models:
* **GPT-5-mini**
* **GPT-4o-mini**
* **Gemini Flash Lite**
* **Mistral Small**
* **Grok 4 Fast**
You can manually select any standard model at any time if you prefer a specific option. Upgrading your plan restores premium access immediately.
***
## Switching Models Mid-Conversation
You can switch to a different model at any point in a conversation:
1. Click the model name in the prompt bar
2. Select a new model from the picker
3. Send your next message — it will use the newly selected model
Switching models does not alter earlier messages. The full conversation history is preserved and passed to the new model as context. Each AI response in the thread displays a small indicator showing which model generated it.
***
## Provider Filter
Use the **provider filter** in the model picker to narrow down options by the company that built the model. Available provider filters include:
* Anthropic · OpenAI · Google · Mistral · DeepSeek · xAI · Perplexity · Qwen · Cohere · Groq · OpenRouter · Kimi · Venice · TheSys · ZAI · Inception · ByteDance · Liquid
***
## Pinning Favorite Models
Click the **star icon** next to any model name in the picker to pin it. Pinned models appear at the top of the model list in a "Favorites" section, making it faster to switch between the models you use most.
***
## Available Providers
ZeroTwo aggregates models from 19 providers:
| Provider | Notable Models |
| ---------- | -------------------------------------- |
| OpenAI | GPT-4.1, GPT-5, o3, o4-mini |
| Anthropic | Claude Sonnet 4.6, Opus 4.6, Haiku 4.5 |
| Google | Gemini 2.5 Pro/Flash, Gemini 3.1 |
| xAI | Grok-3, Grok-4 |
| Mistral | Mistral Large, Mistral Small |
| DeepSeek | DeepSeek latest, DeepSeek R-series |
| Cohere | Command R+ |
| Perplexity | Sonar Pro |
| Qwen | Qwen Max |
| Groq | Fast Llama 3.3 inference |
| OpenRouter | Hundreds of community models |
| Kimi | Moonshot models |
| Venice | Privacy-focused models |
| TheSys | TheSys models |
| ZAI | ZeroTwo-native models |
| Inception | Inception models |
| ByteDance | Doubao models |
| Liquid | Liquid AI models |
***
## Setting a Default Model
To avoid selecting a model manually every time you open a new chat:
1. Open **Settings** (gear icon, bottom-left of sidebar)
2. Navigate to the **Preferences** tab
3. Choose your preferred default model from the dropdown
4. The setting saves automatically — all new conversations will open with that model
Use a fast standard model as your default for everyday browsing, then switch to a premium or reasoning model when you have a task that specifically demands it.
***
## Next Steps
Learn how to enable Web Search, Deep Research, Canvas, and more alongside your chosen model.
Attach documents, images, audio, or video to any chat message.
Compare plan quotas and upgrade to unlock more premium messages.
Read per-provider documentation for model-specific details and capabilities.