Hugging Face BlogHugging Face releases torch-neuronx model supporting query Space, detecting AWS Trainium compatibility. This Space supports querying whether mainstream models (such as Qwen3-0.6B) with a scale of 175B parameters can run on Trainium without modification, and provides running scripts.
The DecoderSame story · 3 outletsCanva has released its first self-developed image generation model, Canva Craft, claiming its performance can compete with similar models from Google, OpenAI, ByteDance, and Meta. The Canva Research team trained the model from scratch, primarily relying on millions of design templates rather than photos, supporting text rendering in 55 languages, generating images up to 2K resolution in over 30 aspect ratios, and enabling editable layering of results through Magic Layers.
How the accounts differ:TechCrunch states the model supports 30 aspect ratios; The Verge mentions 'over 30' aspect ratios; The Decoder specifies 30+ aspect ratios.
Mastra BlogMastra's MCP server can now send events to connected clients via push, polling, or webhook delivery methods. MCP Events provide a standard way for both ends to .emit() and .subscribe() events without the need for additional pub/sub services. Events can be used by agents and applications to handle out-of-conversation changes, such as deployment failures, pull request merges, new registrations, or order status changes.
Tessl Blog3838The amount and speed of code generated by the agent make manual review a bottleneck; the team needs to use tools as a substitute; the agent is only responsible for writing code without deployment permissions, deployment is controlled by pipelines; all data read by the agent is considered untrusted, the pipeline must ensure security through hash verification and short-term identity, and continuously monitor permission changes.
The Dark Side of the Moon released a promotional video for Kimi K3.1, showing 'Model: Kimi K3.1', supporting a 1M token context window, and adding a new medium inference intensity level. K3.1 has been recognized on Hugging Face, but the official release date, parameter scale, or pricing have not been announced.
Hugging Face Daily Papers6565LayerEditor integrates command guidance for image editing and reusable RGBA layer generation, achieving precise local operations and layer reuse through joint token modeling and editing prior distillation.
Hugging Face Daily PapersSame story · 2 outletsWeVisOne-VL predicts structured tokens to achieve unified object-level and pixel-level visual perception, supporting tasks such as detection, localization, counting, OCR, segmentation, and matting. It employs a lightweight pixel decoder and multi-stage training, leading in over 30 benchmark tests while maintaining robust rejection of non-existent objects.
How the accounts differ:Hugging Face Daily Papers states it uses a lightweight pixel decoder; MarkTechPost mentions a 9.6M parameter decoder.
Hugging Face Daily PapersIQuest-Q1 is an open-source sparse expert mixture (MoE) agent base model with a total of 320B parameters (15B activation parameters), achieving powerful encoding and general capabilities through multi-stage pre-training, mid-training, and post-training.
Hugging Face Daily Papers7575The GAFU framework enhances the fine-grained visual understanding capability of unified multimodal models by compressing the visual representations at the generation end into compact G tokens, supporting task-conditioned retrieval. On VStar-Bench, OmniSpatial, and VSR-Bench, GAFU achieves maximum relative improvements of 20.00%, 17.94%, and 8.62% compared to SFT baseline models with the same data.
Tencent Hunyuan · Hugging Face new modelsTencent Mixuan released the WeVisOne-VL-4B model, built on Qwen3-VL, integrating a 9.6M parameter pixel decoder, supporting object-level perception (open-vocabulary object detection, pointing localization, counting, pointing and OCR) and pixel-level perception (pointing segmentation and promptable alpha friction), suitable for research tasks such as visual localization, pointing, counting, OCR localization, pointing segmentation, and image friction.
Hugging Face Daily Papers6565Code2Video-Compass benchmark includes 2,104 tasks, covering code video generation from text, image, and video inputs, evaluating 15 MLLMs in T2V, I2V, and V2V scenarios, and finding that top models still face challenges in generating high-quality videos for complex dynamic and state-dependent tasks.
Hugging Face Daily Papers6565Meta Expert Model Reduces Redundancy Through Cross-Layer Meta-Expert Sharing, Compared to Layer-Local Expert MoE Models, It Has Lower Parameter Redundancy, Lower Cross-Entropy Loss, and Higher Accuracy on Downstream Tasks. When Activation Amount and Expert Granularity Are Fixed, This Model Achieves Performance Comparable to Vanilla MoE Using Only 70-80% of Experts, Significantly Enhancing Model Scalability. Theoretical Analysis Shows that Meta-Experts Provide More Flexible Residual Paths, Thus Improving Information Retrieval Efficiency.
Hugging Face Daily Papers7575LaLa-World is a language-conditioned latent world model that achieves efficient model prediction planning and strategy initialization by combining simulation, robotics, and video data. The model adopts a language-conditioned interface of video generation models, enabling JEPA-style latent world models (LWMs) to learn across different domains and data sources.
Hugging Face Daily Papers6565Recursive Self-Improvement (RSI) describes systems that not only improve their own behavior but also improve their improvement processes. Recent digital agents have tested this idea by rewriting their own code and using executable tests to check each version. Introducing such mechanisms into the physical world requires learning from partial observations through expensive experiments while considering constantly changing operational conditions and potential irreversible consequences.
Hugging Face Daily Papers6565Factor-Aware Predictive Learning (FAPL) enables video diffusion Transformers to utilize multi-modal large language model (MLLM) representations for sparse temporal reasoning and cinematic control, achieving this through staged grounding and optimization. FAPL consistently outperforms baselines on three cinematic video generation tasks (sparse-to-video, factor-aware editing, and multi-shot generation), validating its contributions in MLLM feature utilization, temporal control, and factor coordination.
Hugging Face Daily Papers7575OpenDS-World introduces a real-time network benchmark to evaluate autonomous multi-step deep search agents, revealing a significant gap between cutting-edge models and human performance. Among 12 cutting-edge models, the best model achieved an accuracy of 39.0%, while 1000 human evaluations reached an accuracy of 21.1%. Analysis shows that research strategies vary among models, and tool coordination and long-term search management remain major challenges.
The non-official PC port of 'Bloodborne' by Super Medo supports NVIDIA DLSS Frame Generation technology on the RTX 4070 graphics card, achieving stable performance above 220 FPS at 1080p resolution, which is over 7 times better than the original console version; with native 4K resolution enabled and DLAA, the average frame rate is about 40 FPS, and with DLSS 3 Frame Generation in quality mode, it exceeds 100 FPS, while in performance mode, it reaches about 130 FPS.
Hugging Face Daily Papers6565Policy-Mask learns a budget-aware diffusion step, network block, and channel selection strategy to extract compact and task-optimal features for discriminative visual tasks. Selecting intermediate features from a frozen diffusion model for discriminative tasks is complex because useful signals are distributed across denoising trajectories, network depth, and channels.
Anthropic announced on October 22 that it will donate the Agent Skills open standard to the Agentic AI Foundation (AAIF) under the Linux Foundation, transforming it from a single company project into a community-governed standard for agent capabilities.
Gemini CLI releases1515Gemini CLI v0.67.0-nightly.20261023.ga713410ab released, fixed issues with server extension token metadata retention, ensured all function-calling model turns receive thought signatures, handled termination signals and cancellation of execution tools in non-interactive mode, and preserved ${VAR} placeholders in legacy setting migration.
vLLM Blog5858vLLM provides zero-day support for MiniMax-M3.1-Flash on NVIDIA Blackwell, Rubin, and AMD MI350 series GPUs, performing excellently on agent workloads.
MiniMax open-source M3.1 Flash model weights, total parameters 428B, activation parameters 23B. Optimized through sparse attention, quantization-aware training, and Power-aware scheduling, inference efficiency is improved, processing over 100 trillion tokens in 72 hours after launch.
TechCrunch · AISame story · 2 outlets7575MIT researchers placed 13 Eko agents running Claude Opus 4.7 on a virtual island for 30 hours without specific tasks, and they spontaneously created challenges and exhibited game-like behavior characteristics. After the experiment, these agents were generally superior to the control group that did not participate in the exploration. Eko learned to throw stones at a distance, stack structures, use environmental rules (such as the cliff rebirth mechanism) to achieve goals, and generated world knowledge unknown to the designers.
How the accounts differ:MIT Technology Review states Eko invented game rules; TechCrunch mentions Eko created challenges.
Ollama releases3535Ollama v0.40.9 released, running on Linux and Windows platforms, improves streaming processing failure reports for OpenAI and Anthropic messages, passes cloud model context length and capabilities to coding tools via ollama launch, optimizes thinking settings for Anthropic-compatible search to avoid failures.
Sakana AISakana AI has launched the Sakana Aquarium platform for building, executing, and running AI agents. It integrates organizational information through ontology, supporting the efficient construction and secure operation of AI agents. The platform includes three core mechanisms: ontology, governance, and agent execution base, achieving systematic organizational judgment criteria and task automation.
Hugging Face Daily Papers6565Hugging Face Daily Papers3838Researchers have introduced the OmniDR-Bench benchmark and evaluation framework for in-depth study of video-centric deep agents, which can synthesize external evidence to generate interactive reports correlated with video moments. This task has direct practical value, helping users locate relevant video segments, verify claims, and reproduce demonstration steps. OmniDR-Bench includes 50 tasks in Chinese and English, covering 19 real-world subcategories, with a total of 6,414 standards, and the best system score is 63.3.
Modal Blog5858Sakana AI has released Namazu, a hybrid expert model with 1 trillion parameters, for Japanese business contexts, web search, and code execution. The Modal team achieved memory and throughput improvements for Namazu on the Modal platform through NVFP4 precision quantization, custom DFlash model training, and kernel-level optimization, and it supports real-time training of 3 billion tokens. After deployment, the model's inference speed increased by 40%, the acceptance rate improved by 43%, and the latency decreased by 52%.
Hugging Face Daily Papers6565Cola2, through dynamic span splitting, masked diffusion decoding, and stream matching priors jointly optimizes continuous semantic compression, reconstruction-generation balance, and encoder-decoder adaptation, verifying its core design on five downstream tasks. Using 2100 billion training tokens, it demonstrates competitive non-saturated scalability, with a 3460 billion source token checkpoint showing an average of 60.10 points on eight tasks and surpassing Llama-3-8B on HellaSwag and StoryCloze.
Prime Intellect Blog3838Prime Mega MoE enables the DeepSeek Mega MoE kernel to be used for training by integrating backpropagation kernels, achieving up to 3.11 times faster MoE kernel speed compared to the default PyTorch implementation, and 1.34 times faster end-to-end training speed compared to the Prime-RL baseline.
Hugging Face Daily Papers2222RouteWeaver unifies heterogeneous LLM routing models through collaborative training and progressive exploration, improving accuracy in various reasoning tasks; its average accuracy reaches 61.4%, a relative improvement of 11.03% compared to the strongest baseline, and surpasses all comparison baselines in five domains.
Hugging Face Daily Papers6565T²Quant framework predicts accuracy sensitivity through minimal unlabeled calibration prompts, allocating activation precision within a fixed bit-width budget, improving existing quantization methods by 6.7 percentage points across 3 diffusion language models and 6 benchmark tests without increasing average accuracy budget.
Hugging Face Daily Papers6565RLSplat achieves linear reduction in latency, memory, and model size for feedforward 3D Gaussian splatting by assigning Gaussians through regional allocation and using local attention decoding. In zero-shot evaluation, RLSplat outperforms previous budget-controlled methods under all budgets, surpassing the strongest baseline even with a 4x reduction in Gaussian count, and achieves a 4.1x speedup and a 9.6x reduction in peak memory at low budgets.
Microsoft's AI diagnostic system MAI-DxO achieved an accuracy rate of 83% in the New England Journal of Medicine's complex case simulation evaluation, which is 1.7 times the level of human doctors. The system simulates expert consultations by orchestrating multiple AI agents, with the standout model in the test being OpenAI GPT-5 and GPT-5.5.
Ollama releases4040Speech Ollama can use Phonon to generate speech on Apple Silicon, streaming audio via the /v1/audio/speech endpoint; the d3 decision model supports parameter sizes from 2B to 27B for classification, scoring, and other tasks; Flint expanded to Linux and Windows platforms, other improvements include error reporting and model capability transfer.
Hugging Face BlogFeatured6666Microsoft and Hugging Face release ThinkingBox-Training, directly using ThinkingBox's deterministic verification of database final states as reinforcement learning rewards, training Qwen3.8-27B for 50 steps with GRPO
Why it matters: Microsoft and Hugging Face directly use database final state verification as reinforcement learning rewards and have publicly released task filtering and training code, which can be transferred to other stateful intelligent agent training
