Model Guide

Odeus is model-agnostic. Rather than locking you into a single provider, Odeus lets you choose which model from which provider to use for each conversation. Each model has different strengths, so we encourage you to test the available models and find what works best for your tasks.

Selecting a Model

• When you start a new chat, choose a model from the model selector in the chat input.
• Each conversation runs on the model you pick. A conversation is bound to one model for its duration, so choose the model that fits the task before you start.
• When the task changes, start a new chat and pick the model that suits it.
• Per-message model selection is also available in chat, so you can target a specific model for an individual request.

If a chosen model is temporarily unavailable, Odeus automatically falls back to a comparable available model so your request still completes. This keeps chat working during provider outages, but it is for availability, not task-based routing.

Available models

Odeus supports models from multiple providers. The model selector always reflects the models your workspace has enabled.

Provider	Family	Notes
Anthropic	Claude (Opus / Sonnet / Haiku)	Strong reasoning, coding, and natural-sounding text
OpenAI	GPT	Broad general-purpose capability
Google	Gemini	Strong long-context handling
Mistral	Mistral	Strong multilingual and coding

DeepSeek support is coming soon. Workspace admins control which models are enabled from the admin model catalog, so the exact list you see depends on your workspace configuration.

Understanding Model Naming Conventions

AI providers follow consistent naming patterns that help you identify a model's capabilities without memorizing specific versions.

Version Numbers = Capability Level

Higher version numbers generally indicate newer, more capable models. When a provider releases a new generation, they increment the major version number.

Pattern	What it means
GPT-5 vs GPT-4	GPT-5 is the newer generation
Claude 4 vs Claude 3	Claude 4 is the newer generation
Gemini 2.5 vs Gemini 2.0	Gemini 2.5 is newer within the same generation

When in doubt, choose the model with the higher version number. It typically has better reasoning, fewer errors, and more capabilities.

Size Indicators = Speed vs Intelligence Trade-off

Providers offer multiple sizes within each model family. Models without size indicators are the most intelligent but may be slower. Models with size indicators trade some capability for speed and cost efficiency.

Indicator	Intelligence	Speed	Best for
No indicator (e.g., "Claude Sonnet")	Highest	Moderate	Complex tasks, important outputs
mini / nano	Medium-High	Fast	Everyday tasks, quick iterations
flash / fast	Medium	Very Fast	High-volume work
haiku (Anthropic)	Good	Very Fast	Simple tasks, cost-sensitive use cases

Pro tip: Start a chat with a faster model for drafts and exploration, then begin a new chat on a full model for your final output.

Provider Tiers

Each provider organizes their models into tiers:

| Tier       | Examples      | Use Case                            |
| ---------- | ------------- | ----------------------------------- |
| **Opus**   | Claude Opus   | Most intelligent, complex reasoning |
| **Sonnet** | Claude Sonnet | Balanced intelligence and speed     |
| **Haiku**  | Claude Haiku  | Fast, efficient for simpler tasks   |



| Tier          | Examples            | Use Case                             |
| ------------- | ------------------- | ------------------------------------ |
| **Flagship**  | GPT-5, GPT-5.x      | Most capable, best for complex tasks |
| **Efficient** | mini, nano variants | Fast, cost-effective                 |



| Tier      | Examples     | Use Case                     |
| --------- | ------------ | ---------------------------- |
| **Pro**   | Gemini Pro   | Most capable, complex tasks  |
| **Flash** | Gemini Flash | Fast, real-time applications |



| Tier         | Examples      | Notes                       |
| ------------ | ------------- | --------------------------- |
| **Large**    | Mistral Large | Strong multilingual, coding |

Choosing the Right Model

By Task Type

Task	Recommended Model Type	Why
Quick questions, brainstorming	Fast/mini variants	Speed matters, good enough quality
Writing emails, documents	Standard flagship	Good balance of quality and speed
Complex analysis, research	Flagship model	Need highest accuracy
Coding and debugging	Anthropic Sonnet	Strong at structured tasks
Creative writing	Anthropic models	Known for natural, authentic tone
Long documents	Google Gemini	Excellent long-context handling

Our Recommendations

For Everyday Tasks

Use a current flagship model from Anthropic, OpenAI, or Google. These provide the best balance of capability and speed for general use. Look for models without size indicators (no "mini", "fast", etc.).

For Coding and Writing

Anthropic's Sonnet models are consistently praised for natural-sounding text and strong coding capabilities, which works well for professional communication.

For Complex Reasoning

Choose a flagship model when you need maximum accuracy on analytical tasks.

For Speed-Sensitive Tasks

Flash, mini, or nano variants deliver good results much faster. Useful for iterating on ideas or processing high volumes.

Your model choice also affects how quickly you use your plan allowance. More powerful models cost more per interaction. See Usage Limits for details.

Deep Research model choice

Deep Research is unusual in that you can choose the models it uses per run, across the different roles in a research run, and set the depth (Quick, Standard, or Deep). This is a key Odeus differentiator: most tools lock research to fixed models, while Odeus lets you pick the models for each run.

Staying Current

AI models evolve rapidly. To stay current:

1. Check the model selector - Odeus always shows the models your workspace has enabled.
2. Look for version numbers - Higher numbers generally mean newer capabilities.
3. Try new models - When a new model appears, test it on your typical tasks.

Odeus continuously adds new models as they become available, subject to your workspace's enabled-model configuration.