Search Results for "memory usage" - Page 5

...The model contains approximately 120 billion parameters, but employs a Mixture-of-Experts architecture that activates only a smaller subset of parameters during inference, improving computational efficiency while maintaining high capability. Its architecture combines Transformer attention layers with Mamba state-space components to balance long-context reasoning, memory efficiency, and high-quality language generation. The model is optimized for building AI agents that must perform complex tasks such as planning, tool usage, coding assistance, and multi-step reasoning.

...Parse a video directory to create one image every n seconds, then identify duplicate images and show possible video duplicates for manual analysis. . 1f_parse scan video source folder, replicate with ffmpeg an image folder, then for each image create a fingerprint. . CompareV2 perform the fingerprints comparison and works in memory with constant memory usage. It is folder agnostic and parse fingerprints ordered by source file name. Since 201906 in free Pascal for x3 speed and less memory usage in multi-threading. . 3h_analyse is folder agnostic by first loading in memory current source and image folder. Then a lot of options and cache mechanism enable to get correct performance. ...

...The product supports BYOK (Bring Your Own Key) across major providers, with no markup on API usage. Cross-provider memory uses pgvector with keyword fallback. The workspace ships with 664 prebuilt integrations and 32 languages. Pricing is simple. Free tier exposes the core workspace. Paid plans (Standard $25, Pro $50, Maximum $150) share the same feature set — only credits differ. No per-seat upcharges, no feature gates.