Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Manizales Histórico: Restauración con IA
The talk demonstrates how generative AI restores and animates 1900‑1930 Manizales photographs, detailing the technical workflow, challenges, historical accuracy, and ethical considerations.
Esta charla presenta un proyecto experimental que utiliza inteligencia artificial para restaurar y animar fotografías históricas de Manizales tomadas entre 1900 y 1930. A través de herramientas de IA generativa, se reconstruyen escenas del pasado con el objetivo de explorar nuevas formas de preservar y visualizar la memoria colectiva. Se explicará el proceso técnico detrás de la restauración, los desafíos enfrentados y las reflexiones sobre la precisión histórica y el potencial creativo de estas tecnologías.
- UpscaylUpscayl: The free, open-source desktop AI image upscaler that transforms low-resolution photos into sharp, high-quality visuals using models like Real-ESRGAN.Upscayl is a powerful, free, and open-source desktop application designed to upscale low-resolution images using advanced Artificial Intelligence models. It utilizes architectures like Real-ESRGAN, delivering professional-grade image enhancement by eliminating blurriness and pixelation with a single click. The software offers native cross-platform support for Linux, macOS, and Windows, ensuring consistent performance across all major operating systems. Key features include GPU acceleration, batch processing for high-volume workflows, and support for custom AI models, making enterprise-level image quality accessible to all users without cost.
- ChatGPTOpenAI's Generative Pre-trained Transformer (GPT) model: a conversational AI chatbot for instant text generation, coding assistance, and complex problem-solving.Launched by OpenAI in November 2022, ChatGPT is a state-of-the-art conversational AI, built on the Generative Pre-trained Transformer (GPT) architecture (e.g., GPT-4). The system is fine-tuned using Reinforcement Learning from Human Feedback (RLHF) to produce human-like dialogue, admit mistakes, and reject inappropriate requests. Users leverage the chatbot to execute diverse tasks: generating code snippets, drafting professional emails, summarizing technical documents, and even creating original images via DALL-E integration. It functions as a powerful, multi-purpose tool for rapid content creation and information retrieval.
- GeminiGoogle's natively multimodal AI model: understands and operates across text, code, audio, image, and video.Gemini is Google's most capable and general AI model, engineered from the ground up to be natively multimodal: it seamlessly understands and combines information across text, code, audio, image, and video inputs. The technology is optimized for flexibility, running efficiently on everything from data centers to mobile devices. It is deployed in three key sizes: Ultra (for highly complex tasks), Pro (for broad scaling), and Nano (for efficient on-device tasks). Developers access this power via the Gemini API to build next-generation applications.
- LTX-2LTX-2 is the DiT-based, open-source AI foundation model for production-grade video, delivering native 4K resolution at 50 FPS with perfectly synchronized audio.This is LTX-2: Lightricks' next-generation audio-video foundation model engineered for professional workflows. It's built on a DiT architecture and is the first model to generate video and audio in one coherent, synchronized process. The tech supports native 4K resolution and up to 50 FPS performance, allowing for high-fidelity, continuous clips up to 20 seconds long. We've optimized it for efficiency (up to 50% lower compute cost) and creative control, offering features like multi-keyframe conditioning, IC-LoRA control models, and full open-source weights for maximum customization.
- NanoBananaNano Banana (Gemini 2.5 Flash Image) is the state-of-the-art AI model for rapid, conversational image editing with unmatched character consistency.Nano Banana, powered by Google's Gemini 2.5 Flash Image API, is a next-generation AI image editor. This technology excels at complex, multi-turn creative workflows: simply upload an image and describe your desired edits (e.g., 'place the creature in a snowy mountain'). It delivers flawless results in seconds, operating up to 8x faster than previous models. Key capabilities include superior character consistency, multi-image fusion for seamless composition, and high-resolution output (up to 4K in the Pro version). The model leverages Gemini's deep world knowledge for precise visual reasoning and complex diagram interpretation, making it ideal for professional and commercial projects.
Related projects
Plataformas de IA agéntica de código abierto
Manizales
La charla aborda la instalación, uso y análisis de plataformas de IA agéntica de código abierto como OpenManus,…
De la pereza a la automatización
Manizales
Explores practical automation methods learned over years, showing how simple tools and AI can streamline everyday projects and…
Objetos de aprendizaje conducidos por IA
Medellín
This talk covers developing interactive virtual learning objects using a ReactJS library enhanced with AI to create dynamic…
IA generativa en acción: de datos crudos a insights ejecutivos en horas
Santiago
Se presentarán resultados de la Encuesta Global de Adopción de GenAI, mostrando cómo procesar datos y generar análisis…
IDEAR
Bogotá
The talk explains a web app delivering interactive 3D product models, its analytics pipeline, infrastructure choices, and compares…
Modelos en local, inferencia y algo más
Manizales
Explore deploying and optimizing LLMs locally using Ollama and Groq. Learn quantization, memory optimization, and batching for efficient…