<h1>Choosing an AI model</h1><blockquote>In a nutshell: Tresor offers several AI models. Use Instant mode for fast, direct answers, or Thinking mode for problems that require deep reasoning. All models run inside secure enclaves.</blockquote><h2>Available models</h2>Tresor currently offers these AI models, all running inside confidential computing environments:<table style="min-width: 50px"><colgroup><col style="min-width: 25px"><col style="min-width: 25px"></colgroup><tbody><tr><th colspan="1" rowspan="1">Model</th><th colspan="1" rowspan="1">Strengths</th></tr><tr><td colspan="1" rowspan="1">Kimi K2.5 (Recommended)</td><td colspan="1" rowspan="1">Versatile, strong reasoning, vision capable</td></tr><tr><td colspan="1" rowspan="1">GPT-OSS 120B</td><td colspan="1" rowspan="1">Large, powerful general-purpose model</td></tr><tr><td colspan="1" rowspan="1">DeepSeek v3.2</td><td colspan="1" rowspan="1">Fast, efficient, great for drafting</td></tr><tr><td colspan="1" rowspan="1">Llama 3.3 70B</td><td colspan="1" rowspan="1">Open-source, strong at following instructions</td></tr><tr><td colspan="1" rowspan="1">Mistral 24B</td><td colspan="1" rowspan="1">Fast and uncensored</td></tr><tr><td colspan="1" rowspan="1">GLM 5</td><td colspan="1" rowspan="1">Large context window, multilingual</td></tr></tbody></table>All models support Thinking mode and image understanding. Tresor handles these capabilities at the platform level, so you don't need to worry about which model supports what — just pick the one that fits your task.<blockquote>💡 Tip: If you're not sure which model to use, start with Kimi K2.5 — it's Tresor's recommended model and handles most tasks well.</blockquote><h2>Instant vs. Thinking mode</h2>Models that support reasoning have two modes:<h3>Instant mode</h3><ul class="accent-bullets"><li>Best for: Quick questions, drafting, summarizing, translating, everyday tasks.</li><li>How it works: The model responds directly and quickly.</li><li>Example: "Translate this email to French" or "Summarize this paragraph."</li></ul><h3>Thinking mode</h3><ul class="accent-bullets"><li>Best for: Complex analysis, multi-step problems, strategic planning, legal reasoning, math.</li><li>How it works: The model takes extra time to reason through the problem step by step before answering. You'll see a "Thinking…" indicator while it works.</li><li>Example: "Analyze the tax implications of restructuring our subsidiary" or "Find the logical flaw in this argument."</li></ul>You can use Thinking mode with any model.<h2>How to select a model</h2><ol class="steps-list"><li>Look at the top navigation bar of your chat.</li><li>Click the model selector (it shows the name of the currently active model).</li><li>Choose your preferred model from the dropdown.</li><li>If the model supports it, toggle between Instant and Thinking mode.</li></ol>[Screenshot: The model selector dropdown in the top navigation showing available models and the Instant/Thinking mode toggle]You can change your model at any time, even in the middle of a conversation. The new model applies to your next message.<h2>How to choose the right model</h2>Here are some rules of thumb:| Task | Recommended model | Mode | | ------------------------------------- | -------------------------- | -------- | | Quick Q&amp;A, drafting, emails | DeepSeek v3.2 or Kimi K2.5 | Instant | | Complex reasoning, analysis | GPT-OSS 120B or Kimi K2.5 | Thinking | | Analyzing images or PDFs with visuals | Kimi K2.5 | Either | | Following detailed instructions | Llama 3.3 70B | Instant | | Unfiltered / uncensored responses | Mistral 24B | Instant | | Multilingual tasks | GLM 5 or Kimi K2.5 | Either |<h2>Good to know</h2><ul class="accent-bullets"><li>All models run inside secure enclaves. Your privacy is the same regardless of which model you choose.</li><li>Model availability can occasionally change. If a model is temporarily unavailable, Tresor will let you know — simply switch to another.</li><li>You don't need to pick the "right" model to get started. Try a few and see which one suits your work best. You can always switch.</li><li>Thinking mode takes longer but produces more thorough answers. If you need speed, use Instant mode.</li></ul><blockquote>🔒 Privacy: Every model listed in Tresor runs inside a verified confidential computing environment. Your messages never leave the sealed enclave, regardless of which model you choose. The providers themselves (OpenAI, Meta, etc.) do not receive your messages directly — Tresor's enclave infrastructure processes them.</blockquote><h2>Related articles</h2><ul class="accent-bullets"><li><a href="https://tresorai.productlane.com/docs/getting-started/your-first-conversation" target="_blank" rel="noopener noreferrer" class="cursor-pointer box-border border-b hover:border-b-[2px] transition-all duration-75 " style="text-decoration: none; border-color: var(--portal-accent-color);">Your first conversation</a> — Send your first private message.</li><li><a href="https://tresorai.productlane.com/docs/using-tresor/ai-models-and-thinking-mode" target="_blank" rel="noopener noreferrer" class="cursor-pointer box-border border-b hover:border-b-[2px] transition-all duration-75 " style="text-decoration: none; border-color: var(--portal-accent-color);">AI models and Thinking mode</a> — Change models during a conversation.</li><li><a href="https://tresorai.productlane.com/docs/privacy-and-security/verifying-your-privacy" target="_blank" rel="noopener noreferrer" class="cursor-pointer box-border border-b hover:border-b-[2px] transition-all duration-75 " style="text-decoration: none; border-color: var(--portal-accent-color);">Verifying your conversation's privacy</a> — See the proof for yourself.</li></ul>

Getting Started

Tresor offers several AI models with different strengths. Choose Instant mode for quick answers or Thinking mode for complex reasoning — all running inside secure enclaves.