Nvdium167 published todayThis source uses summary mode. Nvdium provides a summary; read the full article on the original site.
TL;DR We introduce NeoMME, a family of 260M and 800M multilingual multimodal encoders. Unlike many generative visual language models, NeoMME does not use a separate pretrained vision tower or a causal language model. A single bidirectional Transformer processes both text tokens and raw image patches, and we train the e
Earlier this year, Software Forgets: Agent Traces Are the Memory made the case that coding agents already produce the record we keep losing. As they search a codebase, try approaches, hit errors, read documentation, and change direction, they leave behind a dense account of not just what changed, but why. While the dia
On 23 August, Surya Narreddi posted a beautiful video of watercolours painted by a language model. The model writes JavaScript through p5.brush, a library that "adds natural drawing tools to p5.js". The video went viral fast, over 1.5M views at the time of writing. The video came with a blog post explaining the trainin
Structured output is one of the most common real-world tasks for LLMs, yet most benchmarks fold it into broader reasoning or extraction scores rather than measuring it on its own. Whether a model reliably returns valid, parseable output in the requested format and shape — schema compliance — is often what decides wheth
Until now, those decisions have run on outdated economics: one bespoke model at a time and months of expert work on each. So teams model the few hundred series where the money is and cover the rest with safety margins, extra inventory, extra headroom, extra tolerance, acted on after the window has closed. That margin i
Today we’re introducing BenchMIRT, a new method for auditing LLM benchmarks at the level of individual prompts—the questions and tasks a model is scored on. A benchmark is usually designed to measure a particular ability, such as safety, general reasoning, or instruction following. But the individual tasks inside it ma
Today, we are releasing the first layer of that effort: @huggingface/kernels, a minimal library for loading and running optimized WebGPU kernels from the Hugging Face Hub, together with an initial collection of 207 kernels at huggingface.co/webgpu-kernels. The collection covers operations used across a wide variety of
Benchmarks decide what gets built. A model that scores well on the Open ASR Leaderboard gets adopted and iterated on, while capabilities the leaderboard does not measure tend not to improve. Much of the recent work on the leaderboard has gone into making the evaluation metrics more trustworthy: Held-out private splits.
Finetuning multi-vector models involves several components: the model itself, datasets, loss functions, training arguments, evaluators, and the trainer class. I'll have a look at each of these components, accompanied by practical examples of how they can be used for finetuning strong multi-vector models. Lastly, in the
Authors: Granite Team, IBM TL;DR: Granite 4.2 is our first family of dense, decoder-only reasoning LLMs, released in three sizes: 3B, 8B, and 30B. Each model is pre-trained from scratch on roughly 15T tokens with a five-phase strategy that extends the context window to 512K tokens, supervised fine-tuned on chain-of-tho
Our latest paper, Quantization-Aware Healing: A Practical Recipe for Recovering Compressed, 4-Bit LLMs, asks a question that the field has mostly left open: once a model has already been through structural compression, not just quantization, how well does that recovery step actually work, and what is the right way to d
gr.Workflow, built right into Gradio, makes the pipeline the interface. You describe your steps as a graph of typed nodes, and Gradio serves a drag-and-drop canvas where every node is runnable and every intermediate result is visible. The same graph is also a REST API and a one-command deploy to Hugging Face Spaces. Th
One reason is that traditional benchmarks overlook many of the conditions and qualities that make voice systems reliable, natural, contextually appropriate, and effective in practice. That's why we recently introduced held-out sets in Real World VoiceEQ, the Open-ASR Leaderboard, and the Far-field ASR Leaderboard: to m
Of course, making AI research accessible requires a powerful search engine, so that humans and agents can quickly find relevant and related work, either through the website or the pwc search CLI command, which agents can use via the Skill. It's important to note that searching for research is not quite the same as sear
Faster inference: up to 3.18 throughput improvement on a GPU and up to 2.87x on-device. Toward on-device agentic inference: cuts function-calling latency by 57% on average for LFM2.5-2.6B Day-one support for llama.cpp and SGLang: LFM-compatible DSpark integration is open-sourced upstream How does DSpark work The decode
Equipping an agent with agentic memory sounds simple: distill lessons from its past work, put them back in context, and more experience should mean better performance. It doesn't always work that way. When we scaled the evaluation to eight models — from a 30B dense model to frontier proprietary systems — one finding st
Where a regular embedding model compresses a whole text into one vector, a multi-vector model keeps one vector per token and scores query against document with the MaxSim operator. That preserves token-level matching information that a single vector has to average away, which usually means stronger retrieval at the cos
We built a constraint-aware GPU allocator and benchmarked it against a FIFO scheduler across seven benchmark scenarios. On identical hardware, running identical workloads, GPU utilization rose by as much as 33 percentage points, and priority-weighted output rose in every one of them, by as much as 105%. Nothing about t
Models and datasets on HF hub are growing on a daily basis. Public model repositories grew from 2.43 to 2.96 million over the period, datasets from 711,000 to 1 million, Spaces from 1.00 to 1.44 million. The distribution underneath stays extreme, roughly 85.6% of models have fewer than 200 lifetime downloads, and 1.5%
You have an agent that can already record a demonstration and push it to the Hugging Face Hub. Now you want to run that loop continuously: collect episodes through the day, train a policy on the growing dataset, deploy it, and pull the next batch back to improve it. Run that loop once and every piece works. Run it ever