Most AI image detectors work fine in the lab and fail on social media.
The reason is simple. Real photos get compressed, resized and re-uploaded again and again, and detectors trained on clean images read that damage as a sign of fakery. A real photo that has passed through a few platforms often gets flagged as AI-generated.
AIRealNet-2 is built to fix that. It is the upgrade to our original Modotte/AIRealNet, trained on images collected from platforms like Instagram and X, so it has seen the kinds of quality loss and compression real images go through online.
The training set has 98k images, split 50/50 between AI-generated and real, so the model has no class bias to lean on. Personal photos were excluded by design, including AI-edited ones.
The architecture is unchanged: a SwinV2 Tiny backbone with a binary classification head, about 0.2B parameters, working at 256x256 resolution. It stays fast and easy to deploy for research or production.
I built Seekify, a Python tool that turns web search into clean, LLM-ready input for free.
Collecting sources for an LLM workflow usually means juggling search APIs, scraping pages, and cleaning up messy text. Seekify handles that in one place:
- No search API key needed - Structured results with titles, URLs, and snippets - Optional page text extraction so your model gets real content, not just links
Search the web, collect sources, and feed them straight into your pipeline, so you spend less time gathering data and more time building.
It's open source, and I'd love your feedback, ideas, and contributions.
Introducing Live Subtitles, a Windows tray app that captions anything your PC is playing and isn't butt ugly. live and fully local. It runs NVIDIA's Parakeet-unified 0.6B on your GPU, puts words on screen about 0.36 s after they're spoken, and uses less than one CPU core while it does it.
I built a benchmark for it out of real TV, podcasts, earnings calls and a short film. Parakeet-unified scores 12.1% WER there, close to Nemotron streaming's 11.4%, but with much better punctuation and lower worst-case delay. Once a caption is on screen it never gets rewritten or jumps lines, and the overlay disappears when nobody's talking.
No NVIDIA GPU? It falls back to Nemotron on the CPU, which also covers 30 other languages. It can also write SRT/ASS/VTT files for videos on disk.
Windows only. The GPU engine needs an NVIDIA card with ~3.2 GB of free VRAM. MIT.
Edit: the first version of this post said 13.8% vs 17.0%. The replies caught a scoring bug (numbers weren't normalized), so the benchmark now uses Whisper's English normalizer and the numbers above are re-scored.
Most NSFW classifiers break the second an image touches the internet.
They look great on pristine benchmarks, but in the wild, every social platform aggressively recompresses, downsamples, and degrades images. The moment JPEG or WebP compression artifacts show up, confidence collapses and false positives spike.
SafeScan was built to survive actual platform pipelines. Trained on 34,000 images under almost every major social media compression profile using a Vision Transformer backbone (google/vit-base-patch16-224). Instead of blunt binary filtering, it breaks decisions down across 5 clear categories:
• safe • drawing • sexy • hentai • porn
The result is a moderation model that actually generalizes to real-world internet feeds instead of fragile, uncompressed datasets. Open-weight and available on Hugging Face:
Most deepfake audio detectors are quietly cheating.
They don’t really listen to the speech — they just look at how long the embedding vector is. Once they figure that out, accuracy looks great on paper and falls apart in the wild.
AIRealNet-Audio was built to stop that shortcut. It forces every feature onto the unit hypersphere (twice) so the model can only use direction, not magnitude. Trained on speech from 100+ different TTS and voice-cloning systems, plus real human recordings under heavy compression and noise.
The result is a detector that actually has to learn the artifacts instead of gaming the feature space.
So far I've been pointing it at Markdown Minimap, an Obsidian plugin that adds a scrollable IDE-style minimap to your notes. This week I've been clearing a backlog of user-reported issues on it, with Claude often handling them end to end.
Introducing Inflect-v2, two exceptionally small, open-weight English TTS models at just 3.9M and 9.3M parameters. Both generate speech multiple times faster than real-time on CPU. Despite their size, Inflect-v2 delivers quality that is competitive with much larger lightweight TTS systems, including KittenTTS, Piper, and Supertonic-3.
CPU, CUDA, PyTorch, and ONNX are supported. Apache 2.0.
SKT AI Labs, we are pushing the boundaries of AI architecture and research—and today, we are thrilled to open our doors to the global research community!
We warmly welcome researchers, developers, and AI enthusiasts to join us and contribute to our R&D efforts.
🧪 What You Can Explore:
We invite you to experiment with our WMF (Weight Manifold Fusion) technology. You can test this high-dimensional fusion technique on smaller models to gain a deeper understanding of its behavior and token convergence.
If it works: Fantastic! Share your results with us and contribute directly to the core vision of SKT AI Labs.
If it doesn't work: No problem at all! Your critical feedback is just as valuable to us. Every experiment and anomaly helps us refine this architecture to make it more stable and robust.
We firmly believe that true innovation stems from community collaboration and transparent testing. Let's build the future of advanced AI together. Your ideas, test results, and feedback are always welcome!
You Can Still Research and Development On WMF Only SKT-SURYA-H Model is Dismissed.
We’re excited to release NRS_QWEN_MYTHOS_1M — a powerful reasoning model built on Qwen 3.5 9B! At SKT AI LABS, we’ve supercharged this 9B model with our proprietary Neural Reasoning System (NRS) to deliver next-level performance.
🔥 Why This Model is a Game-Changer: ✅ 100x Reasoning Capacity — Exceptional deep logical thinking and complex problem-solving ✅ 1 Million Token Context — Perfect for massive codebases, long documents, and multi-turn agentic workflows ✅ Advanced Thinking Mode — Native <think> tags for true step-by-step Chain-of-Thought reasoning ✅ Tool-Use Ready — Optimized for Python execution, Web Search, and self-correction ✅ Blazing Fast — Runs smoothly on consumer GPUs like RTX 3090/4090
Whether you’re a developer building coding agents, a researcher working with long-context data, or someone who loves powerful reasoning — this model is built for you.
🚀 Sonic: A lightweight Python audio processing library with tempo matching, BPM detection, time-stretching, resampling & track blending — now with GPU (CUDA) acceleration for 10x speed!
Perfect for quick remixes, batch edits or syncing tracks.
This model has been trained and validated on external datasets to support medical research workflows. It is designed to provide reproducible benchmarks and serve as a foundation for further exploration in healthcare AI.
Key highlights: - Built for medical research and diagnostic study contexts - Validated against external datasets for reliability - Openly available to empower the community in building stronger, more effective solutions
This release is part of my ongoing effort to make impactful AI research accessible through **Modotte**. A detailed blog post explaining the methodology, dataset handling, and validation process will be published soon.
Just did something I’ve been meaning to try for ages.
In only 3 hours, on 10 billion+ tokens, I trained a custom BPE + tiktoken-style tokenizer using my new library microtok — and it hits the same token efficiency as Qwen3.
Tokenizers have always felt like black magic to me. We drop them into every LLM project, but actually training one from scratch? That always seemed way too complicated.
Turns out it doesn’t have to be.
microtok makes the whole process stupidly simple — literally just 3 lines of code. No heavy setup, no GPU required. I built it on top of the Hugging Face tokenizers library so it stays clean, fast, and actually understandable.
If you’ve ever wanted to look under the hood and build your own optimized vocabulary instead of just copying someone else’s, this is the entry point you’ve been waiting for.
I wrote up the full story, threw in a ready-to-run Colab template, and dropped the trained tokenizer on Hugging Face.
We are thrilled to announce the launch of SKT-OMNI-CORPUS-2T, a massive-scale, high-quality dataset designed to power the next generation of Foundation Models (LLMs) from scratch. Developed at SKT AI LABS, this corpus is not just a collection of data; it’s a mission to decentralize high-grade AI training for regional languages and global knowledge.
💎 Key Highlights:
•• Massive Scale: Targeting a multi-terabyte architecture for 2T-level tokenization.
•• Pure Quality: Curated from 500+ Elite Sources
•• Structured for MoE: Perfectly sharded into 3.5GB standardized units (SKT-𝕻 series) for seamless distributed training.
🤝 Open for Collaboration!
We are looking for AI researchers, CUDA engineers, and data scientists to join us in this journey of building Project Surya and the ST-X Series models. Whether it's optimization, custom tokenization, or architecture design—let’s build the future together.
We should really have a release date range slider on the /models page. Tired of "trending/most downloaded" being the best way to sort and still seeing models from 2023 on the first page just because they're embedded in enterprise pipelines and get downloaded repeatedly. "Recently Created/Recently Updated" don't solve the discovery problem considering the amount of noise to sift through.
Slight caveat: Trending actually does have some recency bias, but it's not strong/precise enough.
Introducing Seekify — a truly non‑rate‑limiting search library for Python
Tired of hitting rate limits when building search features? I’ve built Seekify, a lightweight Python library that lets you perform searches without the usual throttling headaches.
🔹 Key highlights
- Simple API — plug it in and start searching instantly
- No rate‑limiting restrictions
- Designed for developers who need reliable search in projects, scripts, or apps
📦 Available now on PyPI:
pip install seekify
👉 Check out the repo: https:/github.com/Parveshiiii/Seekify I’d love feedback, contributions, and ideas for real‑world use cases. Let’s make search smoother together!
🚀 Wanna train your own AI Model or Tokenizer from scratch?
Building models isn’t just for big labs anymore — with the right data, compute, and workflow, you can create **custom AI models** and **tokenizers** tailored to any domain. Whether it’s NLP, domain‑specific datasets, or experimental architectures, training from scratch gives you full control over vocabulary, embeddings, and performance.
✨ Why train your own? - Full control over vocabulary & tokenization - Domain‑specific optimization (medical, legal, technical, etc.) - Better performance on niche datasets - Freedom to experiment with architectures
⚡ The best part? - Tokenizer training (TikToken / BPE) can be done in **just 3 lines of code**. - Model training runs smoothly on **Google Colab notebooks** — no expensive hardware required.
