> ## Content Index > Fetch the complete content index at: https://blog.meganova.ai/llms.txt > Use this file to discover other available public pages before exploring further. # Quick Guide: Optimizing Manta Flash in Your Roleplay Proxy Settings - URL: https://blog.meganova.ai/quick-guide-optimizing-manta-flash-in-your-roleplay-proxy-settings/ - Published: 2025-12-25T02:36:17.000Z - Updated: 2026-10-16T15:34:14.000Z - Description: Kill latency in AI role-play. Learn to optimize MegaNova Manta Flash settings for Janitor AI and SillyTavern to boost immersion today. - Author: MegaNova AI - Tags: #Import 2026-07-09 05:42 In the fast-paced world of **AI role-play**, latency is the ultimate immersion killer. When you are in the middle of a high-stakes narrative, a five-second delay feels like an eternity. This "thinking lag" disrupts the creative flow and pulls the user out of the experience. At [**MegaNova**](https://www.meganova.ai/?utm%5Fsource=blog&utm%5Fmedium=social&utm%5Fcampaign=model%5Fsettings%5Fflash&utm%5Fcontent=citation), we view immersion as our core product, not just a feature. Our **Manta** model family is designed to be the world's first "inference cloud" built specifically for these character-driven moments. To truly optimize your experience, you must understand how to leverage [**Manta Flash**](https://console.meganova.ai/serverless/text/Manta-Flash-1.0?utm%5Fsource=blog&utm%5Fmedium=social&utm%5Fcampaign=model%5Fsettings%5Fflash&utm%5Fcontent=citation) through your proxy settings. ## **Advantage of Pre-Generation Routing** The true "latency killer" in our architecture is the **Adaptive Router**. Most AI platforms use "cascade routing," which tries a small model and then retries with a larger one if it fails. This creates a massive time penalty that we find unacceptable for high-level storytelling. MegaNova utilizes **pre-generation routing**. Our system analyzes your message length and structure *before* the first token is generated to select the optimal tier. This single-pass decision minimizes wait times and ensures a "first-token fast" response. ## **Optimized General Settings for AI Role-play** While every character is unique, these general parameters provide a professional baseline for stability and depth. ### **1\. Temperature: 0.7 to 0.9** We recommend keeping your temperature in this range for the best results. A setting of 0.8 is often the "sweet spot." We justify this because it allows for creative variety without losing the logical thread of the conversation. ### **2\. Top P: 0.9** Setting your Top P to 0.9 ensures the model considers the most likely meaningful words while filtering out low-probability noise. This keeps the **roleplay** grounded and prevents the AI from "hallucinating" irrelevant details during intense scenes. ### **3\. Max Tokens: 500 - 800** For long-form storytelling, you want the model to have enough "headroom" to describe actions and dialogue. We suggest this range because it encourages descriptive prose while staying within the efficient processing limits of the Manta Flash tier. ## **Platform Specific Examples** Integrating these "latency killers" into your favorite platforms is straightforward. MegaNova is fully OpenAI-compatible, ensuring your setup is "disruptive" only to the lag, not your workflow. ### **Janitor AI Configuration** **Janitor AI** users can unlock high-performance roleplay by following these steps: - **API URL**: Set your proxy to https://inference.meganova.ai/v1/chat/completions - **Model ID**: Enter meganova-ai/manta-flash-1.0 to access our balanced performance tier. - **Key Benefit**: Tier 2 users enjoy up to 500 Requests Per Day (RPD) on free models, ensuring your sessions are never cut short. ### **SillyTavern Configuration** **SillyTavern** is the gold standard for custom character management. - **API Type**: Select "Chat Completion" and "Custom (OpenAI-compatible). - **Server Address**: Input the MegaNova inference: https://inference.meganova.ai/v1/chat/completions. - **Instruction Template**: Use the "Sentient Being" prompt. We advise this because it reminds the AI to adapt emotionally and remember past context, enhancing the "Manta" experience. ## **Conclusion: Take the Lead in Your Story** Optimizing your [**MegaNova**](https://www.meganova.ai/?utm%5Fsource=blog&utm%5Fmedium=social&utm%5Fcampaign=model%5Fsettings%5Fflash&utm%5Fcontent=citation) settings isn't just about tweaking numbers; it is about reclaiming your narrative flow. By using [**Manta** **Flash**](https://console.meganova.ai/serverless/text/Manta-Flash-1.0?utm%5Fsource=blog&utm%5Fmedium=social&utm%5Fcampaign=model%5Fsettings%5Fflash&utm%5Fcontent=citation)and the correct proxy configurations, you eliminate the technical barriers between you and your characters. Ready to experience the fastest roleplay in the cloud? Head over to the MegaNova dashboard, update your settings, and start your next adventure with zero lag. > Your characters are waiting—don't keep them hanging. ### What’s Next? [**Sign up**](https://console.meganova.ai/register?referral=e0602d28&utm%5Fsource=blog&utm%5Fmedium=referral&utm%5Fcampaign=blog%5Fmn%5Flaunch&utm%5Fcontent=body%5Flink&ref=blog.meganova.ai) and explore now. 🔍 **Learn more:** Visit[ **our blog**](https://blog.meganova.ai/?utm%5Fsource=blog&utm%5Fmedium=referral&utm%5Fcampaign=blog%5Fmn%5Flaunch&utm%5Fcontent=body%5Flink) and[ **documents**](https://docs.meganova.ai/?utm%5Fsource=blog&utm%5Fmedium=referral&utm%5Fcampaign=blog%5Fmn%5Flaunch&utm%5Fcontent=body%5Flink) for more insights or schedule a demo to optimize your roleplay experience. 📬 **Get in touch**: Join our[ Discord community](https://discord.com/invite/QBTnGurDw4?utm%5Fsource=blog&utm%5Fmedium=referral&utm%5Fcampaign=blog%5Fai%5Flaunch&utm%5Fcontent=body%5Flink) for help or[ Contact Us](https://www.meganova.ai/contact-us?utm%5Fsource=blog&utm%5Fmedium=referral&utm%5Fcampaign=blog%5Fmn%5Flaunch&utm%5Fcontent=body%5Flink).