Skip to content

All AI news

Today 1 items
Jul 21Tue
  1. Google DeepMind76

    Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber

    Google DeepMind released three Gemini models: Gemini 3.6 Flash cuts output token usage by 17% versus 3.5 Flash on the Artificial Analysis Index, priced at $1.50/1M input and $7.50/1M output, with DeepSWE 49% vs 37% and OSWorld-Verified 83.0% vs 78.4%.

    Why it matters: The official article gives concrete pricing, token efficiency, and multiple benchmark numbers for the three new models, and explains the deployment scenarios each targets.

Jul 20Mon
  1. AMD Official Announcements72

    Microsoft to Deploy Next-Gen AMD Instinct and AMD EPYC Processors as the Companies Expand Their Long-Term Strategic Partnership

    AMD announced an expanded long-term strategic partnership with Microsoft, which will deploy the AMD Helios Rackscale Solution at scale on Azure for frontier model inference, its own AI customers, and Azure AI services.

    Why it matters: The official announcement specifies the concrete vehicle and product composition of the partnership, helping readers understand how the two companies divide roles in AI infrastructure and the pace of deployment.

Jul 17Fri
Jul 16Thu
  1. Hugging Face Blog32

    Newer Models, Same Advantage

    DharmaOCR scored 0.925 on a Brazilian Portuguese OCR benchmark, beating the newer Mistral OCR4 (0.798) and Unlimited-OCR (0.7587), confirming the value of domain specialization. The model uses supervised fine-tuning to align with Brazilian Portuguese vocabulary and document structure, then DPO to suppress repetitive or incoherent output, achieving the lowest degradation rate. The article argues that the reliability of generative OCR depends on how concentrated the parameters are within a single domain, not on a larger architecture.

  3. Hugging Face Blog87

    Security incident disclosure — July 2026

    Hugging Face disclosed that this week it detected and contained a production infrastructure intrusion driven end-to-end by an autonomous AI agent system, in which attackers exploited two code execution paths via a malicious dataset to gain access to processing nodes, steal some internal datasets and multiple service credentials, with over 17,000 attack event records examined.

    Why it matters: The official disclosure gives a complete account of an agent-driven intrusion and response, plus a practical lesson for defenders on self-hosting open-source models for forensics.

Jul 15Wed
  1. Hugging Face Blog80

    Welcome Inkling by Thinking Machines

    Thinking Machines has released the open-source multimodal model Inkling on Hugging Face, with about 1T total parameters (975B, a MoE with 41B active), natively accepting image, text, and audio inputs, supporting 1M context, trained on 45 trillion tokens, and shipping BF16 and NVFP4 weights.

    Why it matters: The official blog provides architecture details, VRAM requirements for each tier, and deployment configurations, letting readers evaluate multimodal deployment costs and model selection.

Jul 13Mon
Jul 10Fri
  1. Hugging Face Blog61

    Profiling in PyTorch (Part 3): Attention is all you profile

    Hugging Face published part three of its Profiling in PyTorch series, using the profiler on an NVIDIA A100 to compare traces of six attention implementations: naive attention, in-place masking, and SDPA math/efficient/flash/cuDNN.

    Why it matters: The original benchmarks kernel-level differences across six attention implementations with the profiler, teaching readers to read from traces how in-place saves a copy and what each SDPA backend really costs.

Jul 9Thu
Jul 8Wed
  1. Mistral AI49

    Introducing Robostral Navigate

    Mistral has released its first embodied navigation model, Robostral Navigate, an 8B-parameter model built fully in-house and trained on simulation data. Using only a single ordinary RGB camera, with no LiDAR or depth sensors, it can navigate according to natural language instructions.

  2. Hugging Face Blog65

    Native-speed vLLM transformers modeling backend

    Hugging Face announced that vLLM's transformers modeling backend now achieves equal or higher throughput than vLLM's hand-written native implementations for many LLM architectures.

    Why it matters: Official benchmarks show the transformers backend matches or exceeds vLLM's native throughput on three Qwen3 configurations, so model authors no longer need to write custom ports.

Jul 7Tue
  3. Hugging Face Blog61

    Run AI workloads on any cloud, store on Hugging Face: zero-egress storage with SkyPilot

    SkyPilot and Hugging Face jointly released the store: hf storage backend, which mounts Hub repos or Buckets into any SkyPilot task via hf:// URLs; with just an existing HF_TOKEN, data can be read across 20+ clouds, Kubernetes, and on-prem clusters.

    Why it matters: The original gives concrete configuration and measured speeds for zero-egress storage with SkyPilot's cross-cloud mounting, letting readers evaluate storage options for cross-cloud GPU training.

  4. Hugging Face Blog62

    LeRobot v0.6.0: Imagine, Evaluate, Improve

    Hugging Face's blog announces LeRobot v0.6.0, centered on closing the robot learning loop. The new version adds three world model policies (VLA-JEPA, LingBot-VA, FastWAM), a batch of new VLAs (GR00T N1.7, MolmoAct2, EO-1, EVO1, Multitask DiT), and a unified reward model API (Robometer, TOPReward), plus six simulation benchmarks.

    Why it matters: The official release notes fully list the models, benchmarks, and tooling changes, letting readers assess the new version's practical impact on robot learning workflows.

Jul 6Mon
  1. Hugging Face Blog47

    PRX Part 4: Our Data Strategy

    Hugging Face details the data pipeline for its PRX image generation model: mixing public and internal datasets for pretraining, using a VLM to regenerate long captions, then converting them into streamable training data. The text encoder has been switched to Qwen3-VL, with text latents computed online during training at a throughput cost of only about 3–4%. Images are uniformly encoded as JPEG quality 92, and the data is built with Lance and streamed for training via MDS (Mosaic Streaming).

Jul 3Fri
  1. Google DeepMind49

    Google DeepMind and A24 announce first-of-its-kind research partnership

    Google DeepMind and A24 have announced a first-of-its-kind research partnership pairing a frontier AI research lab with a film studio to help artists develop new workflows and techniques. The two will collaborate on long-term R&D across multiple projects: A24's filmmakers can help shape new technologies, while Google DeepMind gets guidance and feedback from working artists. Google has also invested in A24, and the partnership's specific goals and technical outputs will evolve over time.

Jul 2Thu
  1. Mistral AI65

    Leanstral 1.5: Proof Abundance for All

    Mistral released the Apache-2.0 open-source model Leanstral 1.5, with 119B total parameters and 6B activated parameters, trained via mid-training, supervised fine-tuning and CISPO reinforcement learning.

    Why it matters: The official post discloses the full training pipeline, benchmark figures and real code verification cases, letting readers gauge how practical formal verification models are for mathematical proofs and open-source bug finding.

Jul 1Wed
  1. Berkeley AI Research33

    2026 BAIR Graduate Showcase

    The Berkeley AI Research (BAIR) lab releases its 2026 PhD graduate showcase, highlighting graduates such as Baifeng Shi, Charlie Snell, and Hanlin Zhu and their work in robotics and embodied intelligence, LLM reasoning and test-time scaling, generative models, AI safety and human-AI interaction, and speech and medical AI.

  2. Hugging Face Blog61

    Hugging Face and Cerebras bring Gemma 4 to real-time voice AI

    Hugging Face and Cerebras jointly demonstrated a real-time speech-to-speech architecture: voice input is transcribed by Nvidia's Parakeet, reasoning is handled by Google DeepMind's Gemma 4 31B running on Cerebras, and speech output is synthesized with Alibaba's Qwen3TTS.

    Why it matters: The article lays out a complete open-source speech-to-speech stack with components at each layer, so developers can build or swap parts of their own real-time voice AI pipeline.

Jun 30Tue
  2. Google Research61

    Introducing TabFM: A zero-shot foundation model for tabular data

    Google Research has launched TabFM, a foundation model for tabular data classification and regression that frames tabular prediction as an in-context learning (ICL) problem, producing predictions in a single forward pass without manual training, hyperparameter tuning, or feature engineering.

    Why it matters: The article explains TabFM's architecture, synthetic-data training approach, and TabArena benchmark results, letting readers weigh it against traditional tuning workflows.

Jun 27Sat
Jun 25Thu
  1. Google Research47

    Optimizing cloud economics with linear elastic caching

    Google Research's linear elastic caching approach, published at CIDR, models page eviction as a ski rental problem, using a lightweight decision tree model to predict TTLs for cached pages and dynamically resize caches. In Spanner production, it cut memory usage by 15.5% versus fixed-size caches while cache misses rose only 5.5%, lowering TCO by about 5% with only a 0.5% impact on actual I/O costs.