Browse applications built on modern technologies. Explore PoC and MVP applications created by our community and discover innovative use cases for modern technologies.
Generates accurate captions for short videos in four distinct styles—formal, sarcastic, humorous-tech, and humorous-non-tech—using AI-powered video understanding to create engaging, context-aware summaries with consistent formatting and tone.
Framewise watches a video clip and writes accurate captions in four styles — formal, sarcastic, humorous-tech, humorous-non-tech — via a vision-first Fireworks pipeline with grounding and a benchmarked LoRA fine-tune.
A premium AI-powered video captioning pipeline that generates highly contextual, multi-voice subtitles and narrated audio tracks.
Watches a clip and writes four captions in four different voices: formal, sarcastic, tech humor, casual humor. All four come from the same fact-checked scene description, using Gemma 4 through Hugging Face.
A containerized video captioning agent that generates four distinct caption styles — formal, sarcastic, humorous_tech, humorous_non_tech — per clip using Fireworks AI vision models, built for AMD Hackathon Act II Track 2.
The only video captioning studio that QCs itself. Gemma locks the facts, then writes four calibrated voices — formal, sarcastic, tech & non-tech humor — each caption pre-judged for accuracy and tone before you see it. Gemma-first on Fireworks AI.
FourFaced captions video clips in four voices — formal, sarcastic, tech-humor, everyday-humor — using Gemma 4 for the entire pipeline, not just the write-up. Every caption ships with the real evidence behind it, not a confidence score.
TextSink watches a video & speaks in 4 perfect tones. Gemma 4 grounds every clip in visual facts, then writes formal, sarcastic, humorous-tech and humorous-non-tech captions. Plus live styled closed captions and two AIs arguing about what they're watching
AIMRoyale is an AI-powered mathematical arena that generates highly personalized, real-time math challenges based on the user's school grade and difficulty level. Powered by AMD and Framework AI model.
Echo is an automated video processing and captioning pipeline that generates contextual captions across four distinct style tracks, automatically evaluated by a local LLM-Judge for accuracy and tone.
A Dockerized agent that analyzes video clips and generates captions in four modes: formal, sarcastic, tech humor, and everyday humor.
An AI agent that watches short video clips and writes captions in four tones (formal, sarcastic, humorous-tech, humorous-non-tech). A 3-phase pipeline separates factual scene understanding from style writing, built on AMD's Fireworks AI infrastructure.
A video-captioning agent powered exclusively by Gemma 4, grounding every caption in visual evidence and generating four distinct writing styles for each clip.
Kimi K2.6 scene describe per clip (M3 fallback on timeout/deadline), then four parallel DeepSeek rewrites with JSON mode. Overlap, prefetch, and pipelined judge+retry; ~250s projected for 12 clips under a 520s budget within the 10-min cap.
bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla bla
CaptionCaptain transforms videos into engaging captions in multiple styles, helping creators quickly generate content for different audiences. Upload a video or share a link to receive ready-to-use captions in seconds.
A Docker-based batch pipeline that downloads videos, extracts frames with OpenCV, and generates captions in four distinct styles (formal, sarcastic, tech-humorous, and everyday-humorous) using a vision AI model via the Fireworks AI API.
Give it a clip and it writes four captions formal, sarcastic, humorous_tech, humorous_non_tech. All four come from the same real description of what happened, not four separate guesses, so they stay accurate while sounding nothing alike.
This is an ai agent which gives responses such that, it saves maximum possible tokens per answer or response it gives.
A nostalgic Microsoft Bob-inspired AI desktop companion with OpenAI and Anthropic API key integration that can chat, launch apps, open websites, and automate desktop tasks through a customizable animated assistant.
An agent warehouse for creative development. Based Codey is the foundation for a new kind of developer education—self-directed, AI-augmented, and accessible anywhere.
Simulation-first robotic system performing structured physical tasks such as pick-and-place, sorting, and simple assembly. Designed for repeatable execution under varied conditions, environmental interaction, and measurable performance metrics.
This project contains two Solidity smart contracts for a blockchain-based system: 1) a simple bank contract that allows users to deposit and withdraw USDC (under Arc testnet); 2) an ERC721 NFT shop contract that allows users to mint and manage NFTs.
An AI-based web application using a ResNet101 deep learning model to classify chest X-rays as Normal or Pneumonia. The system helps in early detection of pneumonia by providing accurate and instant predictions.
It creates personalized time capsules by reflecting on their current life, goals, & thoughts. The app generates a unique time capsule containing a reflective summary, future predictions, and a heartfelt letter from the user's past self to future self.
Our chatbot support users with mental health issues by filtering content based on tone.It gauges feelings through speech pace, suggests solutions and builds trust with color psychology.
An AI-driven platform for exploring, creating and protecting original humor. Discover joke trends, author original content, and leverage NLP to ensure authenticity.
Co:Sona is a chatbot that impersonates any film character, figure, or model, providing a unique, engaging, and personalized chat experience for users.