Hugging Face BlogHugging Face introduced how to use TorchNeuron for system-level inference performance analysis on AWS Trainium, capturing clean traces with torch.profiler configured for the Neuron backend, and using key_averages() to identify time-consuming operations.
Hugging Face BlogQwen3-4B's specialized vocabulary list VocabX reduces inference cost by 77% by focusing on the cost of each token rather than frequency, without changing the base model. This method for the Portuguese essay scorer reduces the output layer vocabulary from about 152,000 lines to about 65,000 lines, significantly reducing the computational overhead of token splitting and unused vocabulary items.
NVIDIA BlogFeatured6262NVIDIA expands its family of open models to cover language, robotics, biology, and quantum computing, including the new Nemotron 3.5 Super VL and Nemotron 3.5 Bolt.
Why it matters: NVIDIA has rolled out open models across four lines: language, robotics, biology, and quantum, allowing readers to judge the coverage of its open-source ecosystem based on this.
Hugging Face Blog5858NVIDIA launched SimReady Central Library on Hugging Face, aggregating over 2000 OpenUSD assets ready for direct simulation, covering robotic arms, grippers, warehousing equipment, racks, and cooling facilities, free to use without application.
Hugging Face Blog5858FairStress benchmark evaluated LLM performance in demographic fairness across 300 decision scenarios across 29 fields (such as recruitment, healthcare), including 13 million test cases. The study found that the same persuasive prefixes (adversarial prefixes) have varying impacts on model decisions across different demographic groups, and there are differences in how models explain decisions containing demographic information.
Hugging Face Blog5858ALTK-Evolve uses GPT-4.1 to learn and apply customer service scripts. In the τ²-bench customer service agent benchmark test, it surpasses all retail entries in the retail sector, achieving a score of 67.5% Pass^4, and does not require updating model weights. The model is based on a basic ReAct agent and GPT-4.1. It attempts to analyze tasks and use GPT-4.1 to distill success and failure into short behavioral rules, then adds these guidelines to the agent's system prompts without updating weights or manual annotation.
Hugging Face Blog5858AnyLanguageModel 1.0 is released, supporting local model advancements on Apple platforms for a year. New support added for cloud models such as OpenAI Chat Completions, Anthropic Messages, and Google Gemini. Local model support includes llama.cpp, MLX, and Core ML. Tool calling and guided generation features are expanded. Cloud models have the same functionality as local models, and API keys support dynamic acquisition through closures.
Hugging Face BlogFeatured6262The Hugging Face PyTorch team introduces updates to the MPS backend in PyTorch 2.15, adding the tensor torch.float8_e4m3fn and F.scaled_mm, with FP8 tensors occupying half the space of BF16.
Why it matters: The original text provides specific numbers for the MPS backend in FP8, MoE decoding, and memory usage, which can be used to judge the actual changes in running models locally on Mac.
Hugging Face BlogHugging Face has released TorchTPU, making TPU a native backend device for PyTorch. Transformers, Diffusers, and Kernels can now run directly on TPU. Users only need to select a model from the Hugging Face Hub and specify device_map="tpu".
Tencent Hunyuan · Hugging Face new modelsThe Hy-Memory-30B-A3B-GGUF model quantization file released by Tencent's Huayan uses a hybrid precision quantization scheme with imatrix calibration and MoE perception, achieving efficient local inference on llama.cpp, where the ~8 GB expert-mixed build file size is reduced to about 24.71 GB, with performance only 1.53% lower than Q6_K.
Hugging Face BlogTorchNeuron provides a new native backend for PyTorch on AWS Trainium, with native support for Transformers, Diffusers, and Kernels, making it unprecedentedly easy to run models on Trainium and share models on HF Hub.
LangChain Deep Agents releases1717LangChain Deep Agents released version 0.1.86, adding NVIDIA OpenShell as a sandbox provider, fixed multiple bugs including improved alias matching, accurate reporting of workspace rejection, replacing duplicate thread recovery errors, replacing chat input on recall prompts, displaying expiration prompts in the footer, providing actionable errors when Groq tools are limited, and comparing effective interpreter permissions, etc.
Tencent Hunyuan · Hugging Face new modelsQwen3-8B-DSpine-b16 released by Qwen, employing speculative decoding technology with parallel drafting and deep conditional injection, achieving an average acceptance length score of 10.55 on Qwen3-8B. The model supports a context length of 8192 tokens, uses bfloat16 data type and fa3 attention backend, installation requires git clone and pip install, and runs via sglang serve.
Mistral AI · Hugging Face new modelsMistral AI releases Timestral-123M time series base model, with 123M parameters, using a patch-based Transformer architecture, context length of 32,768 time steps, single prediction covers 1,024 time steps, outputs 21 quantiles. The model achieves optimal accuracy in zero-shot prediction on GIFT-Eval, TIME, and fev-bench evaluations, and is licensed under the Modified MIT License.
Mistral AI · Hugging Face new models
Anthropic NewsroomFeatured6060Anthropic announced it has begun providing a team at the nonprofit organization METR with near-employee internal access to evaluate the risks of its frontier models, and METR will publicly release the evaluation results. Anthropic stated it cannot veto METR's release but may request that information in predefined categories be redacted, and said Anthropic ultimately determines which information falls into these categories.
Why it matters: The original text discloses the specific division of the three-tier evaluation access and the boundaries of METR's publication rights, which can be used to compare with third-party evaluation arrangements at other frontier labs.
Mistral AI · Hugging Face new modelsMistral AI releases Timestral-5M time series base model, the 5M parameter version supports zero-shot prediction, context length up to 32,768 time steps, single forward prediction can cover 1,024 time steps, and outputs 21 quantiles. In GIFT-Eval, TIME, and fev-bench evaluations, Timestral-209M achieves state-of-the-art accuracy for zero-shot prediction, and all Timestral models outperform other models at their parameter scale.
Qwen · Hugging Face new modelsQwen/Qwen-Image-2.1-PE-I2I released, a 7B parameter unified text-to-image generation and editing model, featuring a 32-layer single-stream DiT architecture, supporting the generation of regular or transparent images, editing transparent layers, extracting photo subjects, and capable of editing with 10 reference images.
Qwen · Hugging Face new modelsQwen/Qwen-Image-2.1-PE-T2I released, a 7B parameter unified text-to-image generation and editing model, featuring a 32-layer single-stream DiT architecture, supporting generation of regular or transparent images, editing transparent layers, extracting photo subjects, and enabling local editing through circular, doodle annotations, or independent masks. Its text-to-image prompt rewriting model Qwen3.5-VL 9B can convert multilingual requests into detailed English prompts and recommend aspect ratios.
DeepSeek · Hugging Face new modelsDeepSeek-V4.1-Flash is a 552B parameter multimodal MoE model that supports context up to one million tokens, natively processing images and text and generating text autoregressively. It adopts a Causal Encoder-Decoder architecture, reducing global KV cache usage to 890 bytes/token through technologies like CSA2 and FP4 primary KV cache, a 75% reduction compared to DeepSeek-V4-Flash.
Moonshot AI · Hugging Face new modelsKimi K3 is a 2.8T parameter open-source multimodal intelligent agent model released by Moonshot AI, employing Kimi Delta Attention (KDA) and Attention Residuals (AttnRes) architecture.
MiniMax · Hugging Face new modelsMiniMax-M3-MXFP8 is a native multimodal model with a 1M context window, approximately 428B parameters, and 23B activation parameters, achieving 9x pre-filling and 15x decoding speed improvement through MiniMax sparse attention mechanism (MSA). The model leads in long-horizon agent benchmarks, supporting three reasoning modes, and can be downloaded via Hugging Face and deployed using SGLang, vLLM, or Transformers frameworks.
ByteDance Seed · Hugging Face new models2525ByteDance Seed/TaskMem model release, with 31B parameter scale, using BF16 tensor type, but no inference providers are deployed.
ByteDance Seed · Hugging Face new models2828ByteDance Seed/SimArt model pre-training weights released, including multi-modal large language model (MLLM) weights for generating joint-based 3D object structures (URDF files), and sparse VQ-VAE model weights for encoding and decoding 3D part geometries into discrete latent spaces.
LangChain BlogCompanies like Stripe, Ramp, and Coinbase have independently developed internal coding agents that adopt similar architectural patterns, including isolated cloud sandboxes, curated toolsets, Slack first calls, rich boot context, and sub-agent orchestration.
