> ## Documentation Index
> Fetch the complete documentation index at: https://docs.augmentcode.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Available Models

> The LLMs currently available in Augment and how we use them.

## Current models

Augment uses world-class large language models together with our Context Engine. We currently support the following models:

* Claude Haiku 4.5 by Anthropic
* Claude Opus 4.5 by Anthropic
* Claude Opus 4.6 by Anthropic
* Claude Opus 4.7 by Anthropic
* Claude Sonnet 4 by Anthropic
* Claude Sonnet 4.5 by Anthropic
* Claude Sonnet 4.6 by Anthropic
* Gemini 3.1 Pro by Google
* GPT-5.1 by OpenAI
* GPT-5.2 by OpenAI
* GPT-5.4 by OpenAI
* GPT-5.5 by OpenAI
* Kimi K2.6 by Moonshot AI (hosted on Fireworks and Baseten)
* Prism (Claude + Gemini)
* Prism (GPT + Kimi)

## Prism routing options

Prism lets Augment choose the best-fit model for each request. Instead of locking you into a single model, each Prism option routes within a curated model family based on the task, context, and current system conditions.

* **Prism (Claude + Gemini)** routes between Claude Opus 4.7, Claude Sonnet 4.6, and Gemini 3.0 Flash.
* **Prism (GPT + Kimi)** routes between GPT-5.5, GPT-5.4, and Kimi K2.6.

## Choosing a model

You can select the model directly using the Model Picker in the Augment.

* In VS Code and JetBrains, open the Augment panel and use the model dropdown near the input box to switch models.
* In Auggie CLI, use the `/model` slash command or pass the `--model` flag with the desired model.
* Your selection applies only to Agent in that workspace and can be changed at any time.

If you don't pick a model, Augment will use your last selection or the default set by your organization.

## Model pricing

Augment uses a credit-based pricing system. Different models consume credits at different rates based on their capabilities and costs:

* **Sonnet 4.6**: Baseline credit consumption - latest generation Sonnet, balanced capability for medium or large tasks
* **Sonnet 4.5**: Baseline credit consumption - previous-generation Sonnet model, same pricing as Sonnet 4.6
* **Gemini 3.1 Pro**: \~92% of Sonnet's cost - excels at structural thinking, planning, debugging, investigation, and daily execution. Runs at 2.6x cheaper per message compared to Opus 4.6
* **Haiku 4.5**: \~30% of Sonnet's cost - ideal for quick, simple tasks
* **GPT-5.1**: \~75% of Sonnet's cost - great for medium-complexity work
* **GPT-5.4**: \~72% of Sonnet's cost - advanced GPT generation, excels at computer use and multi-agent orchestration
* **Kimi K2.6**: \~50% cheaper than Sonnet - best for cheap agentic work; smarter than Haiku and roughly 3-4x cheaper than Opus 4.7. By Moonshot AI, hosted on Fireworks and Baseten.
* **Prism**: Variable routed cost — on average 20-30% less than frontier model costs, though actual cost depends on the task. Routes within the selected model family to match each request to the best-fit option.

**Premium models:**

* **Opus 4.5**: \~167% of Sonnet's cost - previous-generation Opus model (legacy)
* **Opus 4.6**: \~167% of Sonnet's cost - previous-generation Opus model (legacy)
* **Opus 4.7**: \~167% of Sonnet's cost - newest and most capable Opus model, excels at long-running tasks and complex reasoning
* **GPT-5.2**: \~133% of Sonnet's cost - enhanced reasoning for complex tasks
* **GPT-5.5**: \~143% of Sonnet's cost - most advanced GPT option, designed for highest-complexity coding, computer use, and multi-agent orchestration

<Info>
  Premium models require a credit-based usage plan and may consume credits at a higher rate than other models.
</Info>

<Card title="Credit-Based Pricing" icon="credit-card" href="/models/credit-based-pricing">
  Learn more about credit costs, see detailed examples, and discover tips for optimizing your credit usage
</Card>

## Feature compatibility

All listed models support the core Augment Agent features:

* Deep code understanding with Augment's Context Engine
* Tool use (integrations and MCP), file edits, and multi-step planning

Some behaviors (e.g., wording or style) may vary slightly between models. We'll continue to refine prompts and guardrails to provide a consistent developer experience.

## Notes and transparency

* We may roll out model updates gradually. If you're part of a staged rollout, different workspaces or teammates may see updates at different times.
* For troubleshooting, you can share the request ID with our support team; they can confirm which model handled a specific request.

If you have questions about model availability or want to participate in early access programs, please reach out via Support.