> ## Documentation Index
> Fetch the complete documentation index at: https://docs.creao.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Models
> Compare the AI models available in CREAO — context windows, capabilities, and cost tiers.
## Overview
CREAO gives you access to multiple AI models from different providers. Choose the model that best fits your task — whether you need maximum intelligence, a large context window, or cost efficiency.
Select a model from the **model dropdown** at the top of the chat interface. Your choice persists per thread, so you can use different models for different conversations.
## Model comparison
All models have full access to the same tools — code execution, web search, image generation, file handling, and all connected skills and integrations.
| Model | Provider | Context | Cache | Cost tier | Best for |
| --------------------- | --------- | ------- | ----- | --------- | ------------------------------------------------------------------------------------------------- |
| **Claude Fable 5** | Anthropic | 1M | Yes | Premium | Anthropic's latest flagship — deepest reasoning and long-horizon agent work (eligible paid users) |
| **Claude Opus 4.8** | Anthropic | 1M | Yes | Premium | Frontier reasoning and coding with a 1M-token context window |
| **Claude Opus 4.7** | Anthropic | 1M | Yes | Premium | Hardest software engineering and long-horizon agent work |
| **Claude Opus 4.6** | Anthropic | 1M | Yes | Premium | Complex reasoning at the same price as Opus 4.7 |
| **Fugu Ultra** | Sakana AI | 1M | Yes | Premium | Multi-agent reasoning for hard, long-horizon work |
| **GPT-5.6 Sol** | OpenAI | 1M | Yes | Premium | Largest GPT-5.6 model - frontier reasoning and agentic coding |
| **GPT-5.6 Terra** | OpenAI | 1M | Yes | Standard | Balanced GPT-5.6 reasoning at lower cost |
| **Claude Sonnet 5** | Anthropic | 1M | Yes | Standard | Best balance of speed and intelligence — the default model |
| **Claude Sonnet 4.6** | Anthropic | 1M | Yes | Standard | Previous-generation Sonnet |
| **Claude Haiku 4.5** | Anthropic | 200K | Yes | Economy | Fast, cost-efficient tasks |
| **GPT-5.6 Luna** | OpenAI | 1M | Yes | Economy | Fastest, most cost-efficient GPT-5.6 model |
| **Gemini 3.1 Pro** | Google | 1M | Yes | Standard | Advanced reasoning with multimodal input |
| **Gemini 3.5 Flash** | Google | 1M | Yes | Economy | Fast, agentic Gemini for high-throughput workflows |
| **GPT-5.5** | OpenAI | 1M | Yes | Standard | Smartest GPT model — frontier agentic coding |
| **GLM 5.2** | Z.ai | 1M | No | Economy | Strong coding at very low cost — great for iterative sessions |
| **MiniMax M3** | MiniMax | 1M | Yes | Economy | Sparse-attention reasoning, fast at very long context |
| **Muse Spark 1.1** | Meta | 1M | Yes | Economy | Meta's multimodal agentic model for tools and vision |
Fugu Ultra delivers output in larger batches rather than token-by-token — it thinks first, then sends a complete response. The chat header shows a notice when Fugu Ultra is selected.
## Choosing a model
Choose **Claude Fable 5** — Anthropic's latest flagship, available to eligible paid users — for the deepest reasoning and long-horizon agent work, or **Claude Opus 4.8** for frontier reasoning and coding. **GPT-5.6 Sol** is available to eligible accounts for frontier GPT reasoning and agentic coding. **Fugu Ultra** (Sakana) is also available on paid plans for hard, long-horizon multi-agent work — it uses deep reasoning with a 1M-token context window and delivers output in batches. **Opus 4.7** and **Opus 4.6** are also available if you have existing prompts tuned for a specific version.
**Claude Sonnet 5** is the default and recommended for most users. It handles coding, analysis, and creative tasks well at moderate cost. **GPT-5.6 Terra** is a balanced GPT option for eligible accounts. **Claude Sonnet 4.6** is still available if you have existing prompts tuned for it.
**GLM 5.2**, **GPT-5.6 Luna**, **Muse Spark 1.1**, and **MiniMax M3** use very few credits per message with strong 1M-token context options. GLM 5.2 is especially good at reasoning tasks; MiniMax M3 excels at very long-context work. These models are great for iterative sessions where you send many messages.
**Claude Haiku 4.5** is the fastest model. Use it for quick questions, formatting, or lightweight code generation where speed matters more than depth.
## Using models via the Developer API
Backend integrations can send an optional `model` field on Developer API run-creating endpoints. Use [`GET /v1/models`](/developer-api/models/list-models) to discover the exact model ids your account can run, or see the [Developer API Models guide](/developer-api/models) for request examples and lifecycle policy.
## Prompt caching
Most models support **prompt caching**, which reduces cost and latency on follow-up messages in the same thread. When caching is active, repeated parts of the conversation (system prompt, earlier messages) are served from cache at a reduced rate.
Caching happens automatically. You don't need to configure anything.
## Credit costs
Credits are deducted based on actual token usage. The cost tier determines how many credits each message uses:
| Cost tier | Credit usage per message | Models |
| ------------ | ------------------------ | ------------------------------------------------------------------------------------------ |
| **Economy** | \~0.1–1 credits | GLM 5.2, MiniMax M3, Muse Spark 1.1, GPT-5.6 Luna, Claude Haiku 4.5, Gemini 3.5 Flash |
| **Standard** | \~1–6 credits | Claude Sonnet 5, Claude Sonnet 4.6, GPT-5.6 Terra, GPT-5.5, Gemini 3.1 Pro |
| **Premium** | \~5–40 credits | Claude Fable 5, Claude Opus 4.8, Claude Opus 4.7, Claude Opus 4.6, GPT-5.6 Sol, Fugu Ultra |
Exact costs depend on the length of your message, the conversation history, and how much the model outputs. You can see your remaining credits in the sidebar.