Search Results for "docker-compose" - Page 61

...It uses adaptive thinking to decide when deeper reasoning is needed and coherent thinking to keep reasoning consistent across tasks and modalities. Nex-N2-mini supports image-text-to-text workflows, explicit reasoning traces, robust function calling, and deployment through Transformers, vLLM, SGLang, Docker, and quantized local apps. It performs strongly across agentic, coding, search, and reasoning benchmarks, including SWE-Bench, Terminal-Bench, BrowseComp, Toolathlon, and GPQA.

...The model is built on Qwen3.5-397B-A17B and is designed as the high-quality counterpart to Nex-N2-mini, trading higher compute needs for stronger reasoning and agent performance. It supports image-text-to-text workflows, explicit reasoning traces, robust function calling, and deployment through Transformers, vLLM, SGLang, Docker, and quantized local apps. Nex-N2-Pro performs strongly across agentic, coding, search, and reasoning benchmarks, including Terminal-Bench, SWE-Bench Pro, BrowseComp, Toolathlon, WideSearch, GPQA Diamond, and GDPval.

...It includes both a web interface built with Next.js and a Telegram bot that offers the same guided experience through chat. Both share a common backend in a monorepo structure. The system is fully self-hosted and designed to run in a single Docker deployment including both web and bot services. Users can download videos and audio in formats such as MP4, WebM, MKV, MP3, and M4A, with server-side processing and progress tracking.

...Lastest ships three diff engines side-by-side — pixel (pixelmatch), structural (SSIM), and perceptual (Butteraugli) — so flaky pixel diffs stop crying wolf. A review dashboard tracks baselines, approvals, and audit history. Run it via Docker on your own infrastructure for unlimited screenshots and full data residency, or run on Lastest Cloud for managed CI runners. Built for solo founders, devs doing QA without a QA team, and SaaS teams that need an Applitools-class alternative without the $699+/mo invoice or the data-exfiltration problem.

...It lets you: Ingest events from webhooks (with support for common providers) and scheduled polls Match events using rules and conditions Run actions such as updating shared fields, sending notifications, forwarding HTTP requests, or browser push notifications Build a modular dashboard with status indicators, charts, clocks, recent activity, and other widgets Unlike heavy monitoring stacks (Prometheus/Grafana) or full workflow engines (n8n), Para-Scope is intentionally simple: Single FastAPI process SQLite database No Docker required It is ideal for people who want a central place to watch services, react to events, and keep operational visibility without the complexity of a full observability platform. Landing Page: https://para-scope.bmd-studios.com Source: https://github.com/MrBlindMouse/para_scope

...It excels in mathematics, science, coding, and multi-turn conversation tasks, rivaling or outperforming larger models in several areas. Deployment is supported via TensorRT-LLM, vLLM, and SGLang, with Docker images and integration guides provided. Open-source under a custom license, it's ideal for researchers and developers seeking scalable, high-context AI capabilities with optimized inference.

...Alerts go to Slack, Discord, Telegram, Microsoft Teams, Google Chat, email, SMS, webhook, PagerDuty, ntfy, Pushover and WhatsApp, with dedupe and flap-suppression so brief blips never page on-call. Public data is available as JSON, an RSS feed and an embeddable SVG badge. Run it yourself with Docker (Postgres and ClickHouse) under AGPL-3.0, with no monitor limits on your own hardware. Manage everything as code through the Terraform provider, or connect an LLM client over the MCP server. A hosted tier is free with no card.

PyShelf is an Open Source python based, ebook server, that does not and never will require a windowing system.