Search: “models”

arXiv:2610.00929v1 Announce Type: cross Abstract: Models specialized for the same task converge to similar behavior, yet the parameter updates that produce it share no common coordinate system, so wei

arXiv:2604.07530v3 Announce Type: replace-cross Abstract: Scaling laws describe how language model capabilities grow with compute and data, but say nothing about how long a model matters once released

spec : support DFlash for HunyuanOCR (#28890) model : add DFlash layer-input taps for HunyuanVL DFlash speculative decoding needs the target graph to expose the residual stream entering each layer (re

https://preview.redd.it/nulsv53o8vqh1.png?width=4500&format=png&auto=webp&s=74765dbd409f4c221640f9f6000a685f6fdbb242 Spent weekend benchmarking the Splash engine (by Incoai) and extending its architec

arXiv:2609.17913v1 Announce Type: new Abstract: Face--voice association models may rely on language or gender cues in the voice rather than on speaker-specific voice characteristics, which can lead to

model : add support for HrmTextForCausalLM (DFM Mimir 1B) (#27625) model : add support for HrmTextForCausalLM (DFM Mimir 1B) HRM-Text runs two transformer stacks (low, high) in an alternating cycle ov

Hi 👋 long time lurker It’s been a few years since the last post about Horse Racing within this sub. A few years ago I was reading a popular science book and came across the story Bill Benter and the s

I can't stand kv cache quantization. Even at q8_0, I can feel the difference. But realistically, when running Qwen3.8-27B-UD-Q4_K_XL on my 32 GB GPU, I only have room for ~170k tokens (with mtp and mm

Some people says terminal bench reflects model intelligence better than the intelligent index. From the look of it, the ranking does seem to reflect how people feel about the open and closed models. F

arXiv:2609.10357v1 Announce Type: new Abstract: Time-series foundation models are evaluated almost exclusively on public archives that predate them, so a strong score cannot be separated from having s

One thing that has bothered me about LLM benchmarks for a while is that most of them are essentially snapshots. A model is evaluated, a score is published, and we tend to talk about that score as if i

arXiv:2609.04476v1 Announce Type: new Abstract: Performance modeling is central to hardware design and software optimization, yet constructing these models requires structured reasoning about computat

We all want more tokens per second but I keep seeing confusion on what to expect for given hardware. For the decoding phase (TG/s) to produce one token all the model weights and KV cache needs to be r

🦙 @ollama is used by 9 million developers and 85% of the Fortune 500, giving co-founder and CEO Jeffrey Morgan (@jmorgan) a unique view into which AI models people are actually using and how that’s ch

Integrating modern generative AI capabilities into Kotlin applications shouldn't require juggling raw HTTP clients or bridging disparate Java libraries. Today, we're excited to announce the 1.0 releas

arXiv:2608.29419v1 Announce Type: cross Abstract: Deep learning architectures are increasingly proposed for patient trajectory modeling in electronic health records (EHRs), yet their advantage over si

arXiv:2608.28439v1 Announce Type: new Abstract: One model passed our fidelity check without ever opening the datasheet. We found it while qualifying models for an internal extraction service: a struct

I keep seeing demos of AI agents building scenes in Blender through BlenderMCP, so I tried it myself. I ran both models locally for this and picked the GLM 5.3 family(Q4 quant) because videos of it do

arXiv:2608.26178v1 Announce Type: new Abstract: There is growing interest in whether language models have stable preferences, for technical, safety, and philosophical reasons. We test 20 language mode

arXiv:2608.21727v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) is deployed to make models better at reasoning tasks, but its side effect on what models will di

arXiv:2608.20918v1 Announce Type: new Abstract: Organizations maintain task-specific adapters for open-weight language models, and each new base-model release forces a migration decision: retain exist

Hey guys, Inco AI shipped DFlash 2 a few days ago with a drafter for Qwen 3.8 27B and a llama.cpp PR. I built the PR and ran it against plain decoding, MTP, the n-gram lookup drafters, and my July DFl

arXiv:2608.12307v1 Announce Type: cross Abstract: Recent work on distillation transfers the capabilities of large models to smaller ones often by updating the latter's parameters, through teacher forc

North Micro Vision Instruct is a 2.4B-parameter open-weight vision-language model with native-resolution image support, released under the Apache 2.0 license. It is designed as a compact foundation fo

arXiv:2608.06564v1 Announce Type: cross Abstract: Quantization is how large language models are actually deployed, and below four bits it is known to hurt. What nobody can say is which of the model's

The intersection of medicine and AI has led to remarkable innovations. However, developers now face the thorny challenge of building robust medical AI tools that have been tested and evaluated on dive

arXiv:2608.04719v1 Announce Type: new Abstract: Agent evaluations tell us that a model picked the wrong tool, but rarely why. We introduce canary tools: diagnostic probe tools planted in an agent's Mo

I released LLM 0.32 this morning, the most significant new version of LLM since the initial launch of the project. The new version includes support for visible reasoning traces, server-side provider t

arXiv:2509.21079v2 Announce Type: replace Abstract: Foundation models have shown remarkable capabilities in various domains, but their performance on complex, multimodal engineering problems remains l

Open letters about AI development I wrote this summary of the past few weeks of open letters as a section of my sponsors-only newsletter but I've decided to share it here as well. Open Weights and Ame

arXiv:2508.02092v3 Announce Type: replace-cross Abstract: Large language models represent significant investments in computation, data, and engineering expertise, making them extraordinarily valuable

Welcome to our latest Gemini Enterprise Agent Platform deep dive, a practical walkthrough where we’ll teach you how to build real-world, production-ready agents starting from step 1. If you haven’t al

https://huggingface.co/swiss-ai/Apertus-v1.5-70B https://huggingface.co/swiss-ai/Apertus-v1.5-8B Apertus 1.5 is a family of 8B and 70B parameter language models designed to advance the state of multil

Full-parameter post-training of trillion-parameter-scale MoE models introduces substantial system-level challenges for large-scale distributed training, including severe memory pressure, non-overlappe

arXiv:2607.11327v2 Announce Type: replace-cross Abstract: Model editing keeps large language models (LLMs) up to date without retraining, but temporal facts expose a limitation of the prevailing locat

We're beginning a limited preview of the GPT‑5.6 series: Sol, our flagship model; Terra, a balanced model for everyday work; and Luna, a fast and affordable model. Terra has competitive performance to

Everyone Operating At The Frontier Satya Nadella, Chairman & CEO, Microsoft, interviewed by @saranormous & @eladgil (No Priors) and @swyx (Latent Space) Crossover special at Microsoft Build 2026. Summ

arXiv:2606.01682v1 Announce Type: cross Abstract: Selecting the best response from multiple small-model samples using a stronger scorer is a simple inference-time strategy, but fails when the small mo

arXiv:2606.01481v1 Announce Type: new Abstract: With the rapid advancements in text-to-image diffusion models, generative video models (T2V models) like Sora can now produce short synthetic videos fro

arXiv:2605.08384v1 Announce Type: new Abstract: In this work, we introduce frozen-encoder model composition, a novel approach to multimodal embedding models. We build on the VLM-style architecture, in

Today, we’re thrilled to announce that Gemini 3.1 Flash-Lite, our fastest and most cost-efficient Gemini 3 series model yet, is now generally available. Designed for ultra-low latency, high-volume tas

In this episode of The Agent Factory, Vlad Kolesnikov and I sat down with Omar Sanseviero from the Developer Experience team at Google DeepMind. We explored the groundbreaking release of Gemma 4: a ne

AI is evolving from answering questions to reasoning and taking action. Companies who want to lead in today’s agentic era require computing infrastructure designed and optimized for these new requirem

arXiv:2604.17494v1 Announce Type: new Abstract: Counterfactual explanations (CFEs) are essential for interpreting black-box models, yet they often become invalid when models are slightly changed. Exis

arXiv:2604.18452v1 Announce Type: cross Abstract: Vision-language modeling is rapidly increasing in popularity with an ever expanding list of available models. In most cases, these vision-language mod

arXiv:2508.15815v3 Announce Type: replace Abstract: Modern large language models (LLMs) are typically trained and deployed using structured role tags (e.g. system, user, assistant, tool) that explicit