Skip to main content

​
Current models

Augment uses world-class large language models together with our Context Engine. We currently support the following models:
  • Claude Fable 5.1 by Anthropic
  • Claude Fable 5 by Anthropic
  • Claude Opus 5.5 by Anthropic
  • Claude Opus 5 by Anthropic
  • Claude Opus 4.8 by Anthropic
  • Claude Opus 4.7 by Anthropic
  • Claude Opus 4.6 by Anthropic
  • Claude Sonnet 5.5 by Anthropic
  • Claude Sonnet 5 by Anthropic
  • Claude Sonnet 4.6 by Anthropic
  • Claude Haiku 5.5 by Anthropic
  • Claude Haiku 4.5 by Anthropic
  • DeepSeek V4.1 Flash 1017 by DeepSeek (hosted on Fireworks and Baseten)
  • DeepSeek V4.1 Flash by DeepSeek (hosted on Fireworks and Baseten)
  • Gemini 4 Argon by Google (hosted on Vertex AI)
  • Gemini 3.8 Flash by Google
  • Gemini 3.7 Flash by Google
  • Gemini 3.1 Pro by Google
  • GLM 5.3 Flash by Zhipu AI (hosted on Fireworks)
  • GPT-6 Astra by OpenAI
  • GPT-6.1 Sol by OpenAI
  • GPT-6 Sol by OpenAI
  • GPT-6 Luna by OpenAI
  • GPT-5.6 Sol by OpenAI
  • GPT-5.6 Terra by OpenAI
  • GPT-5.6 Luna by OpenAI
  • GPT-5.5 by OpenAI
  • Grok 4.7 by xAI
  • Grok 4.6 by xAI
  • Grok 4.5 by xAI
  • Kimi K3 by Moonshot AI (hosted on Fireworks and Baseten)
  • Prism (Augment’s proprietary model ensemble combining Claude and GPT models)
  • Prism (Augment’s proprietary model ensemble combining multiple GPT models)
Claude Sonnet 5.5 is available alongside the standard, effort-selectable Sonnet 5. Legacy plans or organization-specific configurations limited to fixed low-effort Sonnet 5 may not offer 5.5. Sonnet 5 remains available under More where Sonnet 5.5 is offered. Claude Haiku 5.5 has a 100K-token context window. Availability is subject to your organization’s model settings. Gemini 4 Argon is available in eligible workspaces that offer Claude Opus 5.5, subject to organization restrictions. It has a 1M-token context window and supports low, medium, and high reasoning effort, with medium as the default. GPT-6.1 Sol is available in eligible workspaces that previously offered GPT-6 Sol. GPT-6 Sol remains available under More where GPT-6.1 Sol is offered. DeepSeek V4.1 Flash 1017 is available in eligible workspaces that offer DeepSeek V4.1 Flash. It supports images, a 1M-token context window, and low, high, or max reasoning effort (high by default). The original DeepSeek V4.1 Flash remains available; 1017 is its recommended upgrade. You can use eligible OpenAI GPT models and the built-in Prism options without approving provider data retention in your organization’s model settings. Other model access restrictions still apply. Claude Haiku 4.5 is scheduled to retire from model pickers on October 15, 2026 (Pacific time).
On September 15, 2026, DeepSeek V4 Flash, GLM 5.2, and Kimi K2.7 Code were retired from model pickers. Use DeepSeek V4.1 Flash 1017, GLM 5.3 Flash, and Kimi K3 for new work.

​
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 5, Claude Sonnet 5, and Gemini 3.0 Flash.
  • Prism (GPT) routes among GPT-5.6 Sol, GPT-5.6 Luna, and a GPT-5.6 Luna solver tier.

​
Choosing a model

  • In Cosmos, choose the model when configuring an expert or starting a session.
  • In Auggie CLI, use the /model slash command or pass the --model flag with the desired model.
  • Your selection 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. For how models are billed and current per-model rates, see Token-Based Pricing.

​
Feature compatibility

All listed models support the core Augment 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.