Détails

Source

TAO Daily

Publication

27 oct. à 01h36

Lien direct

https://taodaily.io/ditto-ai-sn118-uses-cascade-sn91-to-forecast-ai-credit-usage-and-recommend-plans/

Contenu source (brut)

<p class="wp-block-paragraph">AI credits can disappear faster than expected. One day your usage looks manageable, then a few heavy sessions leave you wondering whether your balance will last until the next reset.&nbsp;</p> <p class="wp-block-paragraph">Checking your current balance tells you what remains, but not what you are likely to need.</p> <p class="wp-block-paragraph"><a href="https://taodaily.io/ditto-sn118-launches-partner-program-to-expand-its-business-reach/">Ditto AI</a> (SN118) is addressing that problem with a new in-chat credit forecasting feature that analyses past usage, estimates future consumption, and recommends a suitable plan.</p> <h2 class="wp-block-heading">From tracking credits to predicting costs</h2> <figure class="wp-block-image size-large is-resized"><img fetchpriority="high" decoding="async" width="1024" height="410" src="https://taodaily.io/wp-content/uploads/2026/10/image-252-1024x410.png" alt="" class="wp-image-25886" style="aspect-ratio:2.5060240963855422;width:624px;height:auto" srcset="https://taodaily.io/wp-content/uploads/2026/10/image-252-1024x410.png 1024w, https://taodaily.io/wp-content/uploads/2026/10/image-252-300x120.png 300w, https://taodaily.io/wp-content/uploads/2026/10/image-252-767x307.png 767w, https://taodaily.io/wp-content/uploads/2026/10/image-252.png 1499w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption class="wp-element-caption">Predict Your Usage, Protect Your Budget</figcaption></figure> <p class="wp-block-paragraph">Users can ask questions directly in Ditto&#8217;s chat interface, such as the following;</p> <p class="wp-block-paragraph">1) where their credits have been going.</p> <p class="wp-block-paragraph">2) whether their current pace will exhaust their balance before the next reset.</p> <p class="wp-block-paragraph">3) which plan best fits their needs.</p> <p class="wp-block-paragraph">Ditto combines billing history with information about the user&#8217;s requirements to estimate future consumption and suggest suitable plans. </p> <p class="wp-block-paragraph">The chat provides a short summary of recent activity, projected usage and plan options, while detailed forecasts, model information, and pricing comparisons are available in expanded results.</p> <p class="wp-block-paragraph">Users can access these insights without building spreadsheets or opening a separate forecasting tool.</p> <h2 class="wp-block-heading">How Bittensor Subnet 91 powers the forecast</h2> <p class="wp-block-paragraph">The feature uses <a href="https://ephemeris.cascade.industries/docs"><strong>Ephemeris</strong></a>, a forecasting API developed by <a href="https://taodaily.io/cascade-moe-targets-a-billion-parameter-time-series-forecasting-scale/"><strong>Cascade</strong></a>, which operates Bittensor Subnet 91 (SN91).</p> <p class="wp-block-paragraph">Ephemeris provides access to multiple time-series forecasting models through a single API, similar to how OpenRouter gives developers access to multiple language models.</p> <p class="wp-block-paragraph">Ditto specifically requests <a href="https://arxiv.org/abs/2510.15821"><strong>Chronos-2</strong></a>, an open time-series foundation model developed by <strong>Amazon</strong>. Instead of processing words, it analyses numerical sequences to estimate how values may change over time.</p> <p class="wp-block-paragraph">Here is how the system works.</p> <p class="wp-block-paragraph"><strong>1) Usage data is prepared:</strong> Ditto converts billing records into time-based usage data, starting with hourly intervals. These intervals widen as more history becomes available, and past usage is recalculated at current prices so forecasts reflect current rates.</p> <p class="wp-block-paragraph"><strong>2) Estimates future consumption:</strong> The model produces a range of possible outcomes, including lower (p10), median (p50) and upper estimates (p90). This helps Ditto account for uncertainty instead of presenting one projection as guaranteed.</p> <p class="wp-block-paragraph"><strong>3) Ditto checks the results:</strong> It tests the model against the user&#8217;s past usage and compares its performance with a simple estimate based on recent spending. Chronos-2 is used only when it clearly performs better and meets stricter requirements for predicting when credits could run out. Accounts with insufficient history fall back to a simpler estimate until there is enough data.</p> <h2 class="wp-block-heading">Privacy and plan recommendations</h2> <p class="wp-block-paragraph">Forecasting requires usage data, but Ditto says it sends only anonymised numerical sequences to Ephemeris while prompts, memories and account details stay private.</p> <p class="wp-block-paragraph">The feature also helps users choose the cheapest plan that meets their stated requirements and estimated 30-day usage. Ditto provides a recommendation, while the final subscription choice is completed through Stripe.</p> <p class="wp-block-paragraph">A few limitations are worth knowing.</p> <p class="wp-block-paragraph">1) Usage analysis is available on every plan, but full forecasts require sufficient usage history and available capacity within a shared daily budget.</p> <p class="wp-block-paragraph">2) Forecast results are cached for the remainder of the UTC day so repeated requests can reuse existing results.</p> <p class="wp-block-paragraph">3) Bring-your-own-key (BYOK) usage is tracked separately.</p> <p class="wp-block-paragraph">Users can read the full explanation in <a href="https://heyditto.ai/blog/predict-ai-credit-usage-with-ephemeris">Ditto&#8217;s blog post</a>.</p> <h2 class="wp-block-heading">Why this is useful for Bittensor</h2> <figure class="wp-block-image size-large is-resized"><img decoding="async" width="1024" height="576" src="https://taodaily.io/wp-content/uploads/2026/10/image-253-1024x576.png" alt="" class="wp-image-25887" style="aspect-ratio:1.8909090909090909;width:624px;height:auto" srcset="https://taodaily.io/wp-content/uploads/2026/10/image-253-1024x576.png 1024w, https://taodaily.io/wp-content/uploads/2026/10/image-253-300x169.png 300w, https://taodaily.io/wp-content/uploads/2026/10/image-253-768x432.png 768w, https://taodaily.io/wp-content/uploads/2026/10/image-253-678x381.png 678w, https://taodaily.io/wp-content/uploads/2026/10/image-253.png 1600w" sizes="(max-width: 1024px) 100vw, 1024px" /></figure> <p class="wp-block-paragraph">Ditto is an AI assistant built around persistent memory across apps and chats, including Telegram, Slack, Discord and MCP clients.</p> <p class="wp-block-paragraph">By integrating Cascade&#8217;s forecasting service, it can offer a specialised capability without building every part of the forecasting stack itself.</p> <p class="wp-block-paragraph">This is a practical example of Bittensor subnets serving different roles in a wider product ecosystem. Cascade&#8217;s SN91 supplies forecasting infrastructure, while Ditto uses it to solve a specific problem for its users.</p> <p class="wp-block-paragraph">As more subnet projects connect their services, the network&#8217;s value can extend beyond individual subnets to the products built with them.</p> <p class="wp-block-paragraph">➛ Users can explore Ditto&#8217;s features and plans at <a href="https://heyditto.ai/">heyditto.ai</a> and <a href="https://heyditto.ai/pricing/">view its pricing</a>.</p>