Loading…
Browse by type
Announcement reads
Announcement posts on Yomu, summarised newest first.
- 1,477 posts
- 20 companies
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Announcement posts
Public AI on Hugging Face Inference Providers 🔥
Hugging Face has integrated Public AI as a supported Inference Provider on the Hugging Face Hub. Public AI operates as a nonprofit, open-source project providing access to sovereign and public models from institutions such as the Swiss AI Initiative and AI Singapore. Its distributed infrastructure combines a vLLM-powered backend serving OpenAI-compatible APIs across partner-donated clusters with a global load-balancing routing layer. Users can access these models through the Hugging Face web UI, Python client SDK, and JavaScript SDK using either direct provider API keys or routed Hugging Face tokens. At the time of announcement, inference through the Public AI provider is free of charge, supported by donated GPU time and advertising subsidies.
`LeRobotDataset:v3.0`: Bringing large-scale datasets to `lerobot`
LeRobotDataset:v3.0 is a new release for lerobot, Hugging Face’s robotics library, addressing file-system limitations encountered when LeRobotDataset:v2 stored one episode per file at million-episode scale. Version 3 packs multiple episodes into shared Parquet and MP4 files, while relational JSON and chunked episode metadata identify episode boundaries and locate the corresponding tabular and visual data. The format separates tabular data, visual data, and metadata; it stores low-dimensional readings in Apache Parquet, concatenates camera frames into MP4 videos, and supports native temporal windowing through delta_timestamps. It also introduces StreamingLeRobotDataset for processing batches from the Hugging Face Hub without downloading the full dataset locally, and provides a one-line conversion utility for datasets in the LeRobotDataset format. The release is planned for lerobot-v0.4.0 and is presented as a step toward datasets with millions of episodes.
Usage Policy update
Anthropic is updating its Usage Policy to reflect product changes, user feedback, regulatory developments, evolving AI risks, and enforcement priorities, with changes taking effect September 15, 2025. The update adds explicit prohibitions on malicious computer, network, and infrastructure compromise while continuing to permit vulnerability discovery with system-owner consent, and provides supplementary guidance for agentic use. It narrows political restrictions from broad bans on lobbying and campaign content to prohibitions on deceptive or disruptive democratic activity and voter or campaign targeting. Law-enforcement wording is clarified without changing the permitted scope, while high-risk requirements are specified for consumer-facing legal, financial, and employment uses, including human oversight and AI disclosure. Anthropic describes the policy as a living document and says it will continue evaluation with policymakers, experts, and civil society.
Visible Watermarking with Gradio
Generative artificial intelligence tools produce images, video, audio, and text that have become increasingly difficult to distinguish from real-world media captures. To improve synthetic media transparency, Hugging Face introduced visible watermarking capabilities directly into the Gradio web application framework. Developers can now overlay visual watermarks on image and video outputs by specifying a single watermark parameter using file paths, open image objects, or NumPy arrays. The implementation also supports QR code watermarks, which can match the visual style of generated media while conveying detailed background information. Furthermore, the Gradio Chatbot component accepts a text watermark parameter that automatically attaches attribution metadata whenever users copy generated responses to their clipboards.
Introducing the Palmyra-mini family: Powerful, lightweight, and ready to reason!
WRITER released three open models in the Palmyra-mini family ranging between 1.5B and 1.7B parameters, designed for efficient inference and specialized reasoning tasks. Built on the Qwen architecture, the release includes the standard palmyra-mini base model alongside two Chain of Thought variants, palmyra-mini-thinking-a and palmyra-mini-thinking-b. In benchmark evaluations, palmyra-mini achieved 52.6% on Big Bench Hard, while palmyra-mini-thinking-a reached 82.87% on GSM8K and palmyra-mini-thinking-b reached 92.5% on AMC23. For palmyra-mini-thinking-b, applying reinforcement learning fine-tuning to an OpenReasoning-Nemotron-1.5B base improved single-shot pass@1 accuracy while reducing sampling diversity and majority@64 performance. The models are available in GGUF and MLX-BF16 quantizations and support inference engines including vLLM, SGLang, TRTLLM, and TGI.
Tricks from OpenAI gpt-oss YOU 🫵 can use with transformers
To support OpenAI's GPT-OSS series of models, the transformers library introduced several performance upgrades that apply across supported architectures. The release integrates zero-build custom kernels downloadable directly from the Hub, reducing external dependency bloat and compilation friction for operations like Liger RMSNorm and MegaBlocks MoE. Native support for MXFP4 quantization groups vector elements into 32-value blocks with shared scales, allowing GPT-OSS 20B to fit in roughly 16 GB of VRAM and GPT-OSS 120B in roughly 80 GB. In addition, transformers incorporates Flash Attention 3 with attention sinks, continuous batching via the generate_batch API for experimentation, and automatic memory pre-allocation to speed up model loading on GPUs.
Fine-tune Any LLM from the Hugging Face Hub with Together AI
Together AI and Hugging Face have introduced an integration enabling developers to fine-tune compatible Hugging Face Hub models directly on Together AI's managed infrastructure. To launch a fine-tuning job via the Python SDK, users provide a base model from Together's catalog as a configuration template alongside the target Hugging Face repository identifier. This base model template dictates GPU allocation, memory configuration, training pipelines, and inference setup for custom models with matching architectures and sizes. The workflow operates bidirectionally, pulling from public or token-authenticated private repositories and optionally pushing completed checkpoints back to the Hub upon completion. Geared toward CausalLM models under 100 billion parameters, the capability enables faster iteration cycles and domain adaptation without custom DevOps infrastructure.
The Region Consolidation Project
Fly.io is consolidating its region map and deprecating 17 regions across North America, Europe, Asia-Pacific, South America, and Africa. The announcement lists 17 deprecated locations, including Madrid, Bucharest, Warsaw, Atlanta, Boston, Denver, Guadalajara, Miami, Phoenix, Querétaro, Seattle, Montreal, Bogotá, Ezeiza, Rio de Janeiro, Santiago, and Hong Kong, and names 18 consolidated regions: seven in North America, five in Europe, four in Asia-Pacific, and one each in South America and Africa. For apps migrated by Fly.io, most moves should be seamless, although applications with region-specific logic must be updated to use the new region. Manual moves generally require changing primary_region in fly.toml, scaling up in the destination with fly scale count, and scaling down in the original region. If volumes are involved, the guidance recommends stopping existing machines, forking volumes into the new region, and creating machines from those forks; support is offered for issues that arise.
mmBERT: ModernBERT goes Multilingual
mmBERT is a massively multilingual encoder model trained on more than 3T tokens across over 1,800 languages to improve upon existing multilingual architectures like XLM-R. Built upon ModernBERT with a Gemma 2 tokenizer, mmBERT employs a three-phase training curriculum consisting of pre-training on 60 languages, mid-training on 110 languages, and a final decay phase covering 1,833 languages. The training pipeline integrates an inverse mask ratio schedule, dynamic language temperature annealing, and TIES merging across three decay variants. Benchmark evaluations demonstrate strong natural language understanding on English GLUE and multilingual XTREME, as well as competitive retrieval performance on MTEB v2 and CoIR. The release includes standard base and small models alongside open training data and checkpoints.
Updating sales restrictions for unsupported regions
Anthropic is updating its Terms of Service to address access by companies from regions where its products are not permitted, including entities operating through foreign subsidiaries. The company says ownership or control from jurisdictions such as China can create legal obligations to share data, cooperate with intelligence services, or support military and intelligence objectives. Under the change, companies or organizations more than 50% owned directly or indirectly by companies headquartered in unsupported regions are prohibited from using the services, regardless of where they operate. Anthropic frames the update as aligning its restrictions with real-world national-security risks and reiterates support for export controls, US energy projects, and rigorous evaluation of AI models for capabilities relevant to national security.
Welcome EmbeddingGemma, Google's new efficient embedding model
Google DeepMind released EmbeddingGemma, a multilingual embedding model with 308 million parameters and a 2048-token context window designed for on-device applications. Based on the Gemma3 transformer backbone, the architecture replaces causal attention with bidirectional attention to function as an encoder, followed by mean pooling and two dense layers producing 768-dimensional vectors. The model incorporates Matryoshka Representation Learning, allowing outputs to be truncated down to 512, 256, or 128 dimensions for reduced memory and storage footprints. Trained on approximately 320 billion multilingual tokens across more than 100 languages, the quantized model operates under 200 MB of RAM. In domain-specific evaluations on the MIRIAD dataset, fine-tuning increased NDCG@10 from 0.8340 to 0.8862, outperforming larger baselines.
Meet Mick Hodgins, Notion’s new EMEA GM
Notion announces Mick Hodgins as its new general manager for EMEA, linking the appointment to a priority of expanding its international footprint; 80% of users are outside the US. Hodgins brings 13 years at Google, where he scaled the EMEA ads business and led teams serving customers in more than 20 EMEA countries across changes including mobile, programmatic advertising, the pandemic, and AI. His vision combines local execution with a unified regional team, including new office hubs in London, Paris, and Munich, continued growth of Dublin HQ, and EU data residency. He aims to support students, startups, and enterprises across EMEA while building high-performing teams and positioning Notion as a go-to solution for knowledge workers.
SAIR: Accelerating Pharma R&D with AI-Powered Structural Intelligence
SandboxAQ released the Structurally Augmented IC50 Repository (SAIR), an open-source dataset containing 5.24 million computationally co-folded 3D protein-ligand structures paired with empirical IC50 binding potency data. The repository addresses training data scarcity in structure-based drug discovery, where over 40 percent of the included target proteins lack experimental structures in the Protein Data Bank. To construct the dataset, engineers executed over 130,000 GPU hours of the Boltz1 co-folding model on 760 NVIDIA H100 processors hosted via NVIDIA DGX Cloud on Google Cloud Platform. Infrastructure optimizations maintained over 95 percent GPU compute utilization, compressing the generation timeline from three months to three weeks. Quality validation using PoseBusters confirmed that 97 percent of the predicted complexes met physical plausibility and chemical sanity standards.
Updates to Consumer Terms and Privacy Policy
Anthropic is updating its Consumer Terms and Privacy Policy for Claude Free, Pro, and Max users, including Claude Code sessions tied to those accounts. Users can choose whether their data is used to improve Claude and strengthen safeguards against scams, abuse, and harmful content; the change excludes Commercial Terms services and API use. New users choose during signup, while existing users must decide by October 8, 2025; accepting immediately applies the updates to new or resumed chats and coding sessions, and the setting remains changeable in Privacy Settings. For users who opt in, retention will extend to five years, compared with 30 days otherwise, while deleted conversations are excluded from future training; Anthropic says it filters or obfuscates sensitive data and does not sell user data.
Amazon Nova Models: Now Available on Heroku
Amazon Nova’s nova-pro and nova-lite models are now available through Heroku Managed Inference and Agents, giving applications a managed way to use them. The integration initially supports text-to-text workloads, while multimodal support is planned for a future release; both models support tool use and offer 300,000-token context windows. Nova-lite targets speed, low latency, and high-volume real-time applications at $0.06 per 1M input tokens and $0.24 output, whereas nova-pro targets accuracy and deep reasoning at $0.80 input and $3.20 output. Heroku provides an OpenAI-compatible /v1/chat/completions endpoint for chat and generation, plus /v1/agents/heroku for built-in Heroku tools and custom MCP tools, with provisioning available in US and EU regions through the add-on or CLI.
Bringing the Cursor Agent to Linear
Cursor has introduced a Linear integration that lets teams delegate issues to @Cursor by assignment or comment mention, then runs a background agent with issue details, comments, and linked references as context. Teams can also define triaging rules that automatically assign selected tasks and review agent diffs and pull requests from Linear. After connecting a Linear team through Cursor’s integrations page, users assign an issue or mention @Cursor; the agent analyzes it and notifies users when the work is complete and a pull request is opened. Cursor says its team uses the workflow for thousands of customer-feedback, feature, and bug tasks, including a product-bug and feedback triage rule. According to the post, the agent now drafts pull requests for thousands of issues and can one-shot many of them.
Heroku AI Expands Model Offering with OpenAI’s gpt-oss-120b
Heroku has made OpenAI’s open-weight gpt-oss-120b available through Heroku Managed Inference and Agents, giving developers an OpenAI-compatible chat completions API for compatible SDKs and frameworks. The model uses a Mixture-of-Experts architecture with 117 billion total parameters, activating 5.1 billion per token and enabling operation on a single 80GB GPU; its Apache 2.0 license permits inspection and fine-tuning. It is designed for agentic workflows involving instruction following, function calling, web searches, and Python execution, while reported benchmarks place it near or above several open-weight models and sometimes OpenAI o4-mini. The add-on provides deployment, secure application access, agent actions, monitoring, and scaling without infrastructure management, with pricing of $0.15 per million input tokens and $0.60 per million output tokens.
Find the intersection of security, AI, IAM, and fun at Oktane
Oktane 2025 will focus on security, AI, and identity, with programming for protecting software systems that use both human and non-human identities. The event is scheduled for September 24–26, 2025, at Caesars Forum in Las Vegas, and will include the Oktane Developer Lounge, Expo Hall activities, lightning talks, documentation feedback games, and Developer Support consultations. Hands-on labs cover securing enterprise AI with the OAuth Cross App Access (XAA) extension and Model Context Protocol (MCP), automating Okta with Terraform, replacing standing administrative access with time-bound requests through Okta Identity Governance and Workflows, and using OAuth 2.0 Demonstrating Proof of Possession (DPoP) tokens in a single-page application. An Okta Workflows community meetup will add networking, drinks, and appetizers, while the broader event offers learning, expert conversations, and application-focused discussions.
Announcing The AI Agent Blueprint
The AI Agent Blueprint is a strategic guide for support, customer experience, and AI transformation leaders navigating how to adopt and scale AI Agents. Its first part, “Launch it,” covers building a business case, deciding whether to buy or build, evaluating an agent with a four-step framework, planning deployment, and continuously training, testing, deploying, and optimizing performance. The second part, “Scale it,” focuses on redesigning customer experience, organizational and system design, and economics through AI-first journeys, new roles and governance structures, and revised metrics, ROI models, and reinvestment strategies. The Blueprint argues that launching an agent is only the beginning: sustainable value requires treating AI as infrastructure and support as a potential growth lever, and it is live now.
Updates to Teams and Auto
Two billing updates will apply at the next renewal after September 15 for monthly and yearly plans. Teams is moving agent usage from fixed request costs to variable request costs, so charges rise or fall with the amount of work performed; a simple syntax question costs less than implementing a full pull request. This change extends the credit system already used by individual plans and unifies billing across tiers. For individuals, Auto will count toward included monthly usage at competitive token rates at the next renewal after September 15, replacing its unlimited status since June 2025; the post says Auto had previously matched premium-model pricing for individuals from December 2023 through June 2025 and for teams since June 2025.
Top topics
- Architecture830
- Performance812
- LLMs648
- Machine Learning617
- Open Source561
- AI Agents542
- AI454
- Developer Experience446
Weekly digest
The best engineering reads, once a week.