←Hugging Face
← All topics

Hugging Face

Models, datasets, open-source tools and the Hugging Face community.

Public reading groups
107
Last 30 days
39
Directory status
Curated topic

LatestUS Department of Justice Opens Inquiry into Nvidia's $12.9 Billion Acquisition of Hugging Face

Latest reading groups

Duplicate reports of the same fact are grouped

  1. 21ic Electronics Network
    AI score76

    US Department of Justice Opens Inquiry into Nvidia's $12.9 Billion Acquisition of Hugging Face

    The US Department of Justice has opened an inquiry into Nvidia's $12.93 billion acquisition of Hugging Face, assessing the deal's impact on AI competition, with the core focus being whether the transaction would give Nvidia control of a fundamental distribution point for AI models and software. Nvidia announced the acquisition on September 3, 2026, with consideration including approximately $11.9 billion in cash and up to $1 billion in employee retention equity incentives, and the deal is expected to close in the first half of 2027. Hugging Face hosts more than 3 million AI models and 500,000 datasets, serving over 18 million developers and 200,000 enterprises.

    Why read thisThe article places the DOJ inquiry within the thread of Nvidia's multi-year antitrust investigation, explaining that regulators' real concern is a single chipmaker gaining control of the model distribution gateway.

  2. Tencent HunyuanFirst-partyResearch
    AI score62

    Tencent Hunyuan Releases Hy-Nomas Technical Report and Open-Sources Hy-Nomas-Bench

    Tencent Hunyuan has released the technical report for Hy-Nomas, an interactive video world model, and has open-sourced Hy-Nomas-Bench for reproducible research. Officially, Hy-Nomas was also used to reproduce Genie 4's examples and a comparison video was released, along with an online demo that requires no waitlist. Earlier introductions stated that Hy-Nomas supports free interaction with text prompts and camera trajectories, long-horizon generation with world memory, lightweight LoRA adaptation for fields such as robotics, and optimization for inference on consumer-grade GPUs.

    Why read thisThe original text provides the technical report, the open-source benchmark, and a comparison video with Genie 4, so readers can judge the reproducibility of this world model.

  3. Baseten Engineering BlogFirst-partyModel releases
    AI score68

    Baseten releases decision model joey-1, with P50 latency of 20ms and $0.01 per million tokens

    Baseten's Base Labs has released joey-1, a 2B decision model that directly outputs a single-token label and calibrated confidence, with a P50 end-to-end latency of 20ms, about 2.5 times faster than elephant-1 (49ms), at a price of $0.01 per million tokens, one-tenth that of mammoth-1. joey-1 is post-trained from Qwen3.5-2B and shares the same data recipe and API with elephant-1 and mammoth-1, so switching requires changing only one line. The model is now available on Baseten Model APIs, with zero data retention by default, and its weights are open-sourced on Hugging Face; Model APIs supports a 16k context window, while dedicated deployments can unlock a 262k context window and support fine-tuned weights.

    joey-1

    Why read thisThe original text provides joey-1's latency, pricing, and the division of labor among the three model tiers, so readers can judge whether decision-making tasks should switch to a smaller model.

  4. Hugging FaceFirst-partyModel releases
    AI score60

    NVIDIA Releases Nemotron 3.5 Transcribe and Granary v2 Data Pipeline

    NVIDIA has released the open-weight model Nemotron 3.5 Transcribe on Hugging Face and ModelScope, integrating transcription, speaker diarization, translation, and audio question answering into a single checkpoint. It supports transcription in 40 languages, translation between 31 languages and English, and speaker labeling in 13 languages, can process recordings up to 1 hour long, and can label up to 8 speakers. The model ranks first among open-weight models on the English leaderboard of the Hugging Face Open ASR Leaderboard with an average WER of 5.33%, and achieves 2.8% AA-WER on Artificial Analysis with approximately 4B activated parameters. The simultaneously released Granary v2 covers 48 languages and 463,062 hours of speech data, and the accompanying NeMo Speech fine-tuning recipe reduced WER on held-out speakers from 22.44% to 12.17% in a medical case.

    Why read thisThe original text provides unified weights for multilingual transcription, speaker diarization, and translation, along with public leaderboards and fine-tuning recipes, making it easy to assess deployment approaches.

  5. TechCrunch AIModel releases
    AI score79

    Mistral Releases Weights for Trillion-Parameter Open Model Mistral Large 4, Focused on Cybersecurity

    Mistral has released the weights for its trillion-parameter open-source model Mistral Large 4 (ML4, nicknamed Le Chonk), which enterprises and government agencies can download and deploy themselves; the weights are already available on Hugging Face and via its API. Mistral says the model's performance is comparable to that of the open-source model GLM-5.3, but it consumes 47% fewer tokens per task, and it ranks first among open-source models on Artificial Analysis's Coding Agent Index. Pierre Stock, Mistral's Vice President of Science, said that open weights mean organizations can deploy locally in their own environments to identify, coordinate, and respond to major security incidents in real time.

    Why read thisMistral has released the weights for its trillion-parameter model and provided a token consumption comparison with GLM-5.3, which can be used to assess deployment trade-offs for open-source frontier models.

  1. Hugging FaceFirst-partyModel releases
    AI score83

    Liquid AI Releases the LFM3 Series of Models, Including 1.7B Text, Vision, Audio, and Omni Versions

    Liquid AI has released the LFM3 series of open-weight models, all branched from LFM3-1.7B-Base, covering text, vision, audio, Omni, and the d1 decision model. LFM3-1.7B-Base adopts a three-hybrid-layer architecture, is pretrained on 22T tokens, and in the mid-training stage over the final 3.5T tokens incorporates more agent trajectories and tool-calling data, with the context window extended to 256K tokens. The company says LFM3-1.7B is about 25% faster than LFM2-1.2B at an 8K context, and scores 38.42 versus 3.02 on WebArena-Verified; LFM3-VL-2B's average score across 62 vision benchmarks rises from 57.6 to 71.4. All models are available for download on Hugging Face, along with a single-GPU fine-tuning notebook.

    Why read thisThe original article presents the architectural changes shared by the six models on a common backbone and on-device benchmark data, which can be used to assess the trade-offs of the small-model specialization approach on long-context and agentic tasks.

  1. MiniMaxFirst-partyModel releases
    AI score79

    MiniMax M3.1 Flash releases open-source weights, 428B total parameters, 23B activated per token

    MiniMax M3.1 Flash weights are now public, with 428B total parameters and 23B activated per token. The company claims substantial improvements in agentic tasks, coding, and cybersecurity: BankerToolBench 76.2, DeepSWE v1.1 73.7%, OSWorld V2 61.7, Agents' Last Exam 29.6, CyberGym 82.2. The model uses MiniMax Sparse Attention and DSpark speculative decoding to improve inference efficiency; the weights are now available on Hugging Face and ModelScope, with API and MCode desktop entry points also provided.

    Why read thisThe original text provides the total and activated parameter counts, scores on multiple agent benchmarks, and inference efficiency optimizations, which can be used to assess its capability positioning and deployment cost.

  1. Rohan PaulModel releases
    AI score76

    Recursive releases Flint, a 40B MoE coding model, at 125+ tokens/s on a MacBook Pro

    Recursive has released its first model, Flint, a 40B mixture-of-experts coding model under the Apache 2.0 license, with weights now open on Hugging Face, built from scratch in under 4 months by the automated research system Eureka Machine. Flint generates code at 125+ tokens/s on a MacBook Pro and reaches 150+ tokens/s on an RTX Spark. On a 48GB MacBook, its average quality is about 81%, compared with about 85% for Qwen3.8 27B, but in 128K-token long sessions it is about 7 times faster than the latter.

    Why read thisThis content provides concrete data on the speed-versus-quality trade-offs of a local coding model, which readers can use to judge the feasible limits of local deployment.

  2. UnslothFirst-partyProduct release
    AI score63

    Unsloth releases the Dynamic GGUF quantized version of Flint, runnable locally on 24GB of memory

    Unsloth has released the Dynamic GGUF quantized version of Flint, which can run locally on 24GB-memory devices, along with a running guide. Flint is Recursive's Apache-2.0 model with 39.8B parameters and 2.5B active, which Unsloth calls the strongest MoE model under 50B parameters, reaching 125 t/s on Mac and 150 t/s on RTX Spark. Recursive co-founder Richard Socher previously introduced that the model was built from scratch in under 4 months by the automated research system Eureka Machine, and the weights are now open on Hugging Face.

    Why read thisUnsloth provides the GGUF quantized version of Flint along with a local running guide, so readers can use it to actually deploy this 39.8B-parameter MoE model on a 24GB-memory device.

  3. Modal Official Engineering BlogFirst-partyProduct release
    AI score68

    Modal Releases LLM Training Library Modal Dojo

    Modal has released Modal Dojo, a training library that provides tuned default recipes, modifiable parameters, and an out-of-the-box metrics observability dashboard, supporting models from 500 million to trillions of parameters. The library is built on open-source frameworks such as Miles and Slime, can load datasets from Hugging Face, and lets agents query run status, logs, and trajectories via a CLI. Axiom Bio says it has used it to launch training of DeepSeek V4.1 Flash with a 64k context, running on 64 B300s.

    Why read thisThe original article details the training library's components, tunable parameters, and the scale measured by customers, so readers can judge the barrier to building their own post-training stack.

  4. Hugging FaceFirst-partyModel releases
    AI score73

    NVIDIA Releases Kumo Relational Relational Foundation Model, Predicts Without Training

    NVIDIA has released Kumo Relational, a pretrained relational foundation model that can make predictions directly on enterprise databases in a single forward pass, without task-specific training, feature engineering, or table flattening. Across 41 tasks on 15 databases—RelBench V1/V2, SAP SALT, and 4DBInfer—it outperforms supervised relational deep learning models and gradient-boosted tree pipelines built with expert feature engineering, achieving an 8% higher MRR than the AutoGluon ensemble on SAP SALT. The model is available through NVIDIA's open-source GPU-native structured-data-models library or NVIDIA NIM, and supports databases with 500B+ rows.

    Why read thisThe original text presents comparison results of the pretrained relational foundation model across 41 tasks and a training-free inference path, which can be used to assess changes in the barrier to entry for structured data modeling.

  2. Hugging FaceFirst-partyProduct release
    AI score67

    NVIDIA Releases Cosmos3-H-Dreams Surgical Generative Simulation Model

    NVIDIA has released Cosmos3-H-Dreams, specialized from the 16-billion-parameter Cosmos3-H-Embodied-Nano for action-conditioned video generation of cholecystectomy. The model is trained with 28 action dimensions, 832 x 480 video, and 10 Hz action alignment, and distillation yields two student checkpoints, four-step and two-step; the two-step version is trained separately, offering faster inference but lower image quality. NVIDIA also provides the FlashDreams streaming runtime and methods for adapting to custom data and controllers, and states that the model is a research simulator, not a validated clinical system.

    Why read thisThe original text provides a complete path from distilling the teacher model to four-step and two-step student models, so readers can understand how real-time surgical video generation balances latency and image quality.

  1. Hugging FaceFirst-partyResearch
    AI score63

    ALTK-Evolve Uses Learned Memory to Put GPT-4.1 Customer-Service Agent Atop the τ²-bench Retail Leaderboard

    IBM Research has released ALTK-Evolve on Hugging Face. It uses GPT-4.1 to distill behavioral guidelines from the agent's own training trajectories and append them to the system prompt, achieving 67.5% Pass^4 on the τ²-bench retail task and surpassing all retail entries in its October 12, 2026 leaderboard snapshot. The method requires no weight updates or human annotation. Retail average accuracy rises from 78.1% to 82.5%, and Pass^4 rises from 57.5% to 67.5%; in the airline domain, Pass^4 rises from 35.0% to 45.0% with GPT-4.1 and to 82.0% with GPT-5.5. The authors say the gains come from more accurate decisions rather than longer conversations; the guidelines slightly reduce the number of dialogue turns, and the added cost is mainly in input tokens.

    Why read thisThe original text provides a Pass^4 comparison for the same model with and without learned memory, showing the actual reliability gains from prompt-level memory.

  3. Hugging FaceFirst-partyProduct release
    AI score62

    PyTorch 2.15 Brings FP8 and Faster MoE Decoding to MPS

    PyTorch 2.15 adds FP8 support to the MPS backend and makes MoE model decoding 5–6.5 times faster than 2.14: on an M3 Ultra with torch.compile, Qwen3.6-35B-A3B goes from 9.4 tokens/s to 61.3 tokens/s, and LFM2.5-8B-A1B goes from 23.6 to 123.6 tokens/s. For Wan2.2's tiled VAE video decoding, peak driver memory drops from about 270 GiB to 8 GiB, and the image corruption issue with Qwen-Image-2.1's VAE encoding on MPS is also fixed with a padding change, raising the encode-decode round-trip PSNR from 10.06 dB to 29.95 dB. This version requires macOS 15 or later and Python 3.11 or later.

    Why read thisThe original article provides specific figures for the MoE decoding speedup and the reduction in video VAE memory, which can be used to gauge the actual changes when running models locally on a Mac.

  4. inclusionai_new_reposFirst-partyModel releases
    AI score69

    English translation is being prepared.

    Ming-3.0-tts Logo

  1. Hugging FaceFirst-partyModel releases
    AI score78

    Hugging Face Releases Beam: A 501B Open Source Model Supporting 1M Context and Self-Hosting

    Hugging Face has released Beam, a 501B-parameter sparse MoE open source model aimed at long-text reasoning and agent tasks with 1M context, activating only about 23B parameters per token. The model uses 52 layers of hybrid attention, with 43 layers using a 512-token sliding window and 9 layers using global attention; the KV cache grows by 36 KiB per token, about one-sixth that of a full-attention approach. The official release provides three checkpoints—BF16, FP8, and NVFP4; NVFP4 can compress the weights from about 1 TB to about 280 GB, and day-0 support is already available in transformers, vLLM, and SGLang.

    Why read thisThe original article presents the architectural trade-offs and multi-precision deployment options for the 501B sparse MoE model, which can be used to assess the hardware requirements for self-hosted long-context inference.