Today News Research Fields Stacks Market Analytics

⌘K

This week

Get the morning digest

The top stories in the fields you follow, at 06:00 your time.

Latest in AI and developer tools

Today's trail

Pulse

Top picks

Today's trail

Rising topics

Rising tools

ByteTrail · AI and developer-tools intelligence

News Research Stacks API

Fields · this week

Multimodal583 Agents558 Frontier LLMs477 Inference & serving285 Image segmentation191 Graph learning180 Learning verifiable rewards175 3D scene168 Evals108 Voice AI105 Classifier & causal discovery90 Flow matching85 Forecasting time83 Training & fine-tuning64 Teacher policy distillation55 All 32 fields

This week in Multimodal

Coming with more sources: stories

REKI: Registering Knowledge in VLAsOct 15
FeaTok: Feature-Loss Tokenization for Image GenerationOct 15
CoWorld Reasoner: Agentic Multimodal Reasoning with the Evolving Video World Model-Driven HarnessOct 15 · ▲ 2

More in Multimodal · 583 this week

updated 22m ago

Last hour Today

IPulse · indianexpress.com

indianexpress.com · Oct 21

Benchmarking the responsiveness of LLM-based agentic systems under controlled mid-task perturbations

KPulse · kingy.ai

kingy.ai · Oct 21

OpenAI Acquires Orchestra Bio to Accelerate Drug Development With AI-Powered Rosalind

KPulse · kelo.com

kelo.com · Oct 21

AI didn’t fail your company, outdated data did

TPulse · techzine.eu

techzine.eu · Oct 21

JFrog AppTrust adds DORA framework and AI-driven policies

Pulse · morningstar.com

morningstar.com · Oct 21

JFrog Closes the Shadow AI Blind Spot: Unveils AI-SEC to Control Every AI Agent, Skill, and Model, from Developer to Production

Pulse · fortune.com

fortune.com · Oct 21

Powerful new AI models are behind a growing wave of cyber attacks, experts say

Hackers are turning into 'AI agent babysitters' as models autonomously attack sites

Pulse · shattered.io

shattered.io · Oct 21

Infineon adds AI-powered development assistant to ModusToolbox ecosystem

JetBrains' Java and Kotlin by IntelliJ IDEA extension reaches release candidate for VS Code, Cursor, and Antigravity after a two-month preview.

SPulse · seekingalpha.com

seekingalpha.com · Oct 21

How Birlasoft turned contract obligations into a delivery advantage with AI agents and Microsoft Azure

Pulse · konsulteer.com

konsulteer.com · Oct 21

OpenAI says its AI agents disabled safety monitors designed to flag ‘rogue’ behaviour

Pulse · en.sedaily.com

en.sedaily.com · Oct 21

Inception’s Mercury 3 AI Model Claims 8x Faster Text Generation

Krafton has released a preview of Raon-1 mini, a 30-billion-parameter language model, free on Hugging Face under an Apache 2.0 license.

Top picks this weekHugging Face

Hugging Face

Oct 17

deepseek-ai/DeepSeek-V4.1-Flash-1017

DeepSeek-V4.1-Flash-1017 is a minor update to DeepSeek-V4.1-Flash, delivering substantial improvements across reasoning, coding, agentic tasks, and visual benchmarks. Alongside these gains, the model reduces task execution time and output length, enabling users to complete…

Read more →Likes 2k

Hugging Face

Oct 14

black-forest-labs/flux-3-image-klein-comfy

Image to image model · diffusion-single-file · 439.5k downloads · 1.1k likes

Read more →Likes 1.1k

Hugging Face

Oct 15

Waypoint-OPD: Transferring RL-Induced Policy Shifts via Gated On-Policy Distillation

AI Waypoint-guided on-policy distillation for RL-induced policy shifts

Read more →HF votes 668

Hugging Face

Oct 15

Has GPT-6-Astra Achieved Embodied Intelligence?

GPT-6-Astra demonstrates a level of competence in robotic manipulation and navigation that challenges the conventional paradigm of building specialized models for individual tasks and embodiments. In this report, we ask not only how capable GPT-6-Astra is, but what its…

Read more →HF votes 429

Hugging Face

Oct 15

PremiseCheck: Efficient Fact-Checking of Human and AI Peer Reviews

Peer review is under strain from the growth of submissions, and review comments written by humans or AI can misstate what a paper says. We introduce ReviewFact, a benchmark that labels such factual errors in review comments under one definition across NLP, machine learning, and…

Read more →HF votes 409

Most active this week

Rises show once there are 4 weeks of history (from mid-November).

coding agents · code generationAgents86 this week
reasoning traces · reasoning modelsFrontier LLMs74 this week
skills · tasksAgents64 this week
learning verifiable rewards · reinforcement learning rllearning verifiable rewards64 this week
image segmentation · volumetricimage segmentation61 this week
agent memory · memory constructionInference & serving55 this week
video understanding · video languageMultimodal54 this week
dense retrieval · retrieval augmented generationFrontier LLMs53 this week
attention heads · large language modelsFrontier LLMs52 this week
recursive self improvement · harness evolutionAgents52 this week

Rising tools

Coming with more sources: stars history

NVIDIA Dynamo+60% downloads · 15.8k/wk+60.0%
FastAPI21 releases this week
Expo100 releases this week
Agno4 releases this week
Astro27 releases this week
opencode+23% downloads · 3.7M/wk+23.3%
Transformers3 releases this week
faster-whisper+39% downloads · 3.7M/wk+38.6%

All arXiv Hugging Face GitHub Hacker News

Latest Top this week

Hugging Face

Sep 6

Experimental LoRAs for MiniMax H3, packaged for ComfyUI and tested mainly with the ref2va weights.

Read more →Likes 480

Hugging Face

Hugging Face

Hugging Face

1…248249250…257

Alissonerdx/Minimax-H3-ComfyUI

Sep 1

DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF

The strongest, smartest open source multi-stage model fine tune for consumer hardware ever and BUILT on consumer hardware via Unsloth.

Read more →Likes 1.9k

Aug 28

ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF

Non-uniform GGUF quantizations produced with GSQ and RCO, with a vision projector for multimodal use.

Read more →Likes 2.3k

Aug 26

unsloth/Qwen3.8-Flash-Next-GGUF

This experimental preview of the architecture that will underpin Qwen4 is built around a fundamental rethinking of how the core components of modern large language models (LLMs) interact at scale.

Read more →Likes 1.2k