World Wire

Supersonic Labs , a small AI lab from Brazil, has released Julia 1 . It is a compact decision model, not a chatbot. You pass it context, a question, and 2 to 20 candidate answers. It picks one and returns a probability for every option. The model has 144.3M parameters and runs on a plain CPU. Is it deployable? Yes. The weights are on Hugging Face under Apache 2.0 and run locally with Python 3.11+ on CPU or a BF16-capable GPU. An ONNX build also runs in the browser via WebGPU. A hosted API is announced but not open yet. What Julia 1 Does Julia 1 handles three decision types through one API: choice: pick one label from 2 to 20 described options (classification, routing). score: return the expected index on an ordered rubric, such as low, medium, high. noul: return the probability that a yes-or-no statement is true. Results come back in the caller’s option order with full softmax probabilities. Caller IDs such as billing are returned unchanged. T...

In this tutorial, we build a comprehensive multimodal augmentation and robustness workflow with AugLy for images, text, and audio. We start by addressing modern dependency compatibility issues and generating deterministic synthetic datasets so the experiments remain self-contained and reproducible. We then explore AugLy’s functional and class-based APIs, metadata, and intensity tracking, probabilistic composition, bounding-box-aware transformations, and custom transforms. We extend the workflow into practical robustness experiments by benchmarking perceptual-hash copy detection under image distortions and evaluating text classifiers against adversarial perturbations, Unicode obfuscation, sanitization, and adversarial training. We also integrate audio augmentation, build a queryable metadata warehouse, and connect AugLy transformations directly to PyTorch datasets and DataLoaders, giving us an end-to-end view of augmentation as both a data-generation mechanism and a measurable ro...

Liquid AI has announced LFM2.5-VL-3B-DSpark , an experimental speculative-decoding draft model for its LFM2.5-VL-3B vision-language model. The drafter adds about 280M parameters and speeds up decoding without changing the model’s output. Liquid AI team reports up to 3.13x faster decoding on Apple silicon and up to 2.66x on an NVIDIA H100. Is it deployable? Yes, Weights are live on Hugging Face in Safetensors and GGUF , with day-one support in SGLang, MLX-VLM, and llama.cpp. Liquid AI team labels the release experimental, and it ships under the LFM Open License v1.0 , which allows free commercial use only for companies under $10M in annual revenue. What Speculative Decoding Changes for a VLM A standard model generates one token per forward pass. Speculative decoding adds a small drafter that proposes several tokens ahead. The large target model then checks the whole block in one pass and keeps the tokens it agrees with. DSpark follows the recipe from Liquid AI...