Handy Computer Shrinks Parakeet Ultra to 486 MB for Python-Free Transcription
Moondream's Parakeet Ultra speech-to-text model now ships as GGUF quantizations for transcribe.cpp, delivering 1.8% WER at 486 MB across 25 languages.
20 stories
Moondream's Parakeet Ultra speech-to-text model now ships as GGUF quantizations for transcribe.cpp, delivering 1.8% WER at 486 MB across 25 languages.
Cadence is an open-source Python library for brains that learn continuously from a live stream without backpropagation, attention, or frozen weights.
A research prototype streams a live 3D Gaussian Splatting model instead of video frames, letting thin clients render their own viewpoints from cloud scenes.
A Y Combinator-backed Swiss lab open-sourced its autonomous security agent harness, which has already surfaced over 30 Linux kernel zero-days.
Zyphra's new open-source TTS model runs real-time voice cloning on phones and laptops with 100M active parameters and Apache 2.0 weights.
A community experiment converts Liquid AI's 24B mixture-of-experts model into a block masked-diffusion sampler, hitting 600 tokens/second on one A100 but with broken outputs.
An open in-context LoRA on Hugging Face reconstructs MiniMax-H3 drafts at 2K or 4K in three to four steps, filling a gap MiniMax left closed.
A new benchmark tests whether multimodal models can use eye-gaze to follow attention, infer intent, and predict actions in live egocentric video.
Google DeepMind researchers show a coordinated swarm of attacker agents still cracks frontier models that human red-teamers struggle to break.
Stanford and ETH Zürich researchers introduce a per-token mixed-policy self-distillation method that cuts supervised tokens by 38 to 69 percent while preserving general skills.
A new research paper from TU Munich shows that cloud rendering can skip path tracing entirely and still match reference quality in about 10 milliseconds.
Stanford researchers show AI scientists invent more original ideas when they get less task information up front, then learn from past experiments to pick winners.
Artificial Analysis rebuilt its Decision Index around a calibration-aware metric, private document benchmarks, and symmetry tests that penalize models for wobbly confidence.
Continuity Research released Tess-4.5-27B, a Qwen3.8-27B fine-tune that matches its base on BenchLocal using 25% fewer tokens and runs locally.
A set of llama.cpp patches and custom quantizations lets a 284B DeepSeek MoE model run entirely in 64GB of system RAM with zero GPUs.
Sim is an open source workspace that bundles a visual agent builder, 270+ integrations, knowledge bases, tables, and run monitoring in one place.
A community distill pours Claude Opus and Sonnet reasoning traces into Liquid AI's 1.7B tri-hybrid model, packaged as GGUF quants that fit in about 1GB.
BeautifulRust bundles three Agent Skills and a tested clippy config that push coding agents toward idiomatic Rust across Claude, Codex, Copilot, Cursor, and Gemini.
Concept Country rewrote nine weight matrices in Qwen3.8-27B with no training, cutting reasoning tokens up to 54% while raising benchmark scores.
IBM's third-generation TinyTimeMixer brings mixture-of-experts routing, trend-residual decomposition, and roughly 15 to 50 times faster forecasting than popular SOTA baselines.