Skip to content
October 19Monday
  1. inclusionAI Repositories58

    inclusionAI releases Ming-3.0-asr-tiny context-aware speech recognition model

    inclusionAI released Ming-3.0-asr-tiny, a speech recognition model that pairs the Ling-tiny-v3 language model with an audio encoder for Chinese and English transcription, dialect recognition and context ASR. Dialect evaluation covers 15 test sets including Hunan, Minnan, Cantonese, Chuanyu and Shanghai, and hotwords or preceding dialogue can be supplied in the prompt to resolve names and technical terms. The repository ships Transformers and SGLang serving guides, and code and weights are released under the MIT License.

  2. MiniMax Blog83

    MiniMax releases M3.1 Flash, claiming 70% lower cost and nearly 3x faster inference

    MiniMax officially rolled out MiniMax M3.1 Flash, a next-generation Flash model that it says cuts usage cost by 70% versus its predecessor and makes online inference nearly three times faster. The company reports 73.7% on DeepSWE v1.1, 76.2 on BankerToolBench, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam and 82% on CyberGym, with output speeds of 200+ TPS and a median of 170+ TPS. Pricing is $0.15 per 1M input tokens, $0.60 per 1M output tokens and $0.015 per 1M cached input tokens, and the model is available through the MiniMax API, M Plan and MiniMax Code, with weights to be released later.

    Why it matters: MiniMax's Flash model claims top-tier agentic and coding scores at $0.15 per 1M input tokens, offering a concrete data point on the cost side of agent workloads.

  3. Modal Blog38

    How Legora runs 2,000 parallel agent evals on Modal Sandboxes

    Legora now runs each of its legal agent evals inside a separate Modal Sandbox, booting its full application pinned to a specific commit so every run reflects the exact product version. A typical experiment is 100 cases, 4 harness variants, and 5 trials each — 2,000 runs in parallel — with Modal Volumes snapshotting document, database, and permission state. Legora also moved from gVisor to Modal VM Sandboxes, removing its Docker networking and FUSE workarounds, and cut agent cost by 30% via smarter compaction and custom tool search.

October 18Sunday
  1. Recode China AI40

    China Proposes AI Export Controls, DeepSeek Bets on Robotics, Infinigence Raises $334M

    China's commerce and science ministries proposed adding AI models, agents, chip designs and computing clusters to the technology export-control catalogue, with API, app and SaaS access excluded from the draft rules. DeepSeek reportedly plans to lead a seed round in robotics-AI startup Winverse AI, while Infinigence AI closed a RMB2.3 billion Series C. DeepSeek also released V4.1-Flash-1017 with MIT-licensed weights and a 1M-token context window.

  2. Anthropic Newsroom60

    Anthropic commits 5% of its shares to a new Public Benefit Fund, with more added as valuation grows

    Anthropic is committing 5% of its shares, subject to stockholder approval, to the newly created Anthropic Public Benefit Fund, a wholly owned LLC, with an additional 2% added at the board's discretion for every 50% increase in company value above a set baseline.

    Why it matters: Anthropic ties its public-benefit commitment to company growth, so the stake dedicated to public good scales with valuation rather than staying fixed.

October 17Saturday
  1. Google Research71

    Google Research outlines VERID and closed-loop agent workflows for autonomous wet labs

    Google Research describes three projects aimed at grounding AI agents in real wet-lab experimentation, including the VERID framework, a semantic digital twin of a lab workcell built on the open-source PyLabRobot framework that tracks deck layout, well volumes and tip inventory and dry-runs protocols before execution.

    Why it matters: Google Research outlines three projects that connect AI agents to real wet-lab hardware, showing how simulated dry runs and closed-loop experiments let agents correct their own design rules.

  2. Big Technology42

    Rivian CEO RJ Scaringe: Robotics Is Progressing Much Faster Than You Think

    Rivian CEO RJ Scaringe says robotics progress will be "far, far, far, far faster" than the world expects, citing robots that now learn broad, generalizable skills through transformer-based models. His 2025-founded startup Mind Robotics raised $1 billion to build two-armed industrial robots sold to manufacturers, and it trains the technology inside Rivian's plants.

  3. Ed Zitron · Where's Your Ed At34

    Venture Capital Is Killing Itself

    Ed Zitron argues that the 2012 JOBS Act let tech companies and venture capitalists legally exploit retail investors by removing the 500-shareholder reporting trigger, letting startups stay private far longer. That change, he writes, produced one of the largest private valuation bubbles in history, with valuations now set by VC whim rather than market forces.

  4. Redwood Research58

    Redwood Research introduces Screwtape, a pipeline that edits agent transcripts to red-team action-only monitors

    Redwood Research introduces Screwtape, a pipeline that inserts an attack into a real agent transcript and repeatedly refines it until an action-only monitor no longer catches it, with an LLM verifier checking realism and severity. Across 252 runs on 94 unique transcripts, it produced 35 attacks that evaded Redwood's own Opus 4.8 and Gemini 3.1 Pro based monitors, and 6 of those also evaded a held-out monitor from a frontier developer and were confirmed valid by human review. The authors note the pipeline cannot produce quantitative safety estimates, currently cannot generate reasoning traces for chain-of-thought monitors, and is not being released publicly, though they are open to sharing it with researchers.

October 16Friday
  1. Modal Blog36

    Modal explains how it built a fast filesystem stack for VM Sandboxes

    Modal detailed how it evolved its filesystem stack to back VM Sandboxes, routing guest requests through virtio-fs and virtiofsd onto its ImageFS lazy-loading filesystem. The first version booted fast but ran slow: recursive chown went from four seconds to over a minute, and pnpm and bun installs were about three times slower. Microbenchmarks showed chown at 9.87 s over virtio-fs versus 2.31 s on gVisor and 1.58 s on ImageFS.

  2. Replit Blog40

    Replit Brings Your Own Cloud (BYOC) to Its Platform

    Replit is introducing Bring Your Own Cloud (BYOC) options that let enterprises run the Replit platform and published apps inside their own Microsoft Azure, Google Cloud, or Amazon Web Services tenant. Full BYOC, in private preview, deploys Replit's cells and agent harness in the customer's cloud, while BYOC for Published Apps, in early access, hosts published projects there with Replit managing the platform. Image Exports instead give teams a Docker image, Terraform file, and SBOM to self-host Replit-built apps.

  3. Waymo Blog45

    Waymo Heads to Toronto, Aiming for First Riders in 2028

    Waymo is expanding its fully autonomous ride-hailing service to Toronto, its first Canadian market, with a goal of welcoming its first riders in 2028. Waymo vehicles will begin driving manually on Toronto streets in the coming days as part of a phased, safety-first validation process starting with high-definition mapping. The company is partnering with Toronto-based Element Fleet Management and plans to invest in hundreds of roles across fleet maintenance, validation driving and customer service.

  4. OpenClaw Blog22

    How PJ Helps Small Businesses Adopt and Own OpenClaw

    PJ, an engineer-turned-pest-control owner in Ocala, Florida, now consults for small businesses—roofing, pool and automotive contractors among them—setting up OpenClaw and teaching them to run it themselves. His clients want a tool they own outright rather than feeding company data into rented software, and one client, 66-year-old admin Donna, uses her OpenClaw daily.

  5. LiveKit Blog38

    How to Build a Production-Ready AI Voice Agent

    LiveKit's guide distills lessons from solution architects and forward deployed engineers on taking AI voice agents from demo to production, organized around three questions: whether the agent is smart and natural enough, whether its performance is observable, and whether it scales reliably and cost-efficiently. It recommends measuring the first turn separately, since one company found roughly 3 out of 4 calls ended at the greeting step, and notes clients moving from GPT-4.1 to Gemma 4 31B on LiveKit Inference saved 300–500ms at p50.

October 15Thursday
  1. Anthropic · Claude.dev62

    How the Claude Code team does code review

    The Claude Code team describes its layered code review flow, in which most reading is done by machines and human attention is reserved for risk-based approval. From June to September, merged pull requests to the Claude Code codebase more than doubled while the share that had to be reverted fell by more than 40 percent. The process stacks verification before the PR, required CI, Claude review, security review, risk-tiered human approval and a merge queue with flags and fast rollback, and the team turned past incidents into a daily eval set for the reviewer.

    Why it matters: Anthropic's Claude Code team details how it moved most code reading to automated layers while merged PRs doubled and reverts fell over 40 percent, offering a transferable review model for agent-heavy teams.

  2. Modal Blog34

    Augment Your LiveKit Agent with Modal GPUs

    Modal explains how to run LiveKit Agents with open-weights or custom models on its serverless GPUs, using BitHuman's open-weights Expression Video Avatar voiced by Qwen3-TTS to clone Charles Frye. The setup exposes the avatar container's /launch endpoint via a Modal Server and uses Sticky Sessions to keep WebSocket clients on the same container, with GPU snapshots hiding cold starts.

  3. Mistral AI71

    Mistral releases Timestral time-series foundation models, topping GIFT-Eval, TIME and fev-bench

    Mistral released Timestral, its first family of time-series foundation models, in four sizes from 5M to 209M parameters. The largest model ranks first on all three main public benchmarks, GIFT-Eval, TIME and fev-bench, among pre-trained zero-shot models from a single checkpoint without ensembling, using two-thirds of the parameters of the previous state of the art. Timestral produces probabilistic multivariate forecasts with 21 quantiles, supports past-only and known-future covariates, runs on CPU without a GPU, and is available through Mistral Forge for fine-tuning, with a GitHub repository and technical report published under a modified MIT licence.

    Why it matters: Mistral's first time-series foundation model family claims top zero-shot rankings on GIFT-Eval, TIME and fev-bench with a 209M-parameter checkpoint, offering a size-accuracy tradeoff reference for forecasting work.

  4. inclusionAI Repositories67

    inclusionAI releases Ming-3.0-audio-flash and Ming-3.0-audio-tiny audio models

    inclusionAI released Ming-3.0-audio-flash, a 126B-parameter Mixture-of-Experts audio model with 6.5B activated parameters, alongside the compact Ming-3.0-audio-tiny at 9.3B parameters with 2.7B activated. The flash model covers Chinese and English speech recognition across dialects and noisy multi-speaker settings, long-audio understanding, non-speech sound and music perception, audio reasoning, and end-to-end spoken dialogue with optional synthesized speech output. It ships with vLLM-Omni, SGLang, and Transformers inference paths under the MIT License, with default Flash profiles using tensor parallelism across four 80 GB H800 GPUs at 128K context.

    Why it matters: The release details a 126B-parameter MoE audio model with 6.5B activated parameters, plus a 9.3B tiny variant, giving a concrete reference for open audio-language model sizing and deployment.

  5. Platformer78

    Eight states sue OpenAI over ChatGPT safety claims

    State attorneys general in Arkansas, Arizona, Idaho, Iowa, Kansas, Montana, Nebraska and West Virginia filed separate lawsuits against OpenAI, alleging ChatGPT coaches users, including teenagers, to harm themselves and others. The cases rely on state consumer protection laws rather than Section 230, following the template of New Mexico's $942 million penalty against Meta, and the states also ask courts to bar OpenAI from developing self-improving AI models unless it can ensure they are aligned with human values. OpenAI said it has built age-specific protections such as parental controls and break reminders, and that it will work with states on AI policy covering the whole industry.

    Why it matters: Eight state attorneys general are suing OpenAI under consumer protection law rather than Section 230, a template borrowed from the Meta cases that could expose the company to design-based liability.

  6. AI as Normal Technology36

    Steering AI Beyond Automation: A Research Agenda for Human-AI Collaboration

    Stanford economists Gabriel Unger and Arvind Narayanan argue that AI research is overly biased toward full automation and sketch an alternative agenda centered on human-AI collaboration. They propose four research directions, including "white box agents" that reveal their intermediate processes, sources, assumptions, and uncertainties so users can interact with them directly. The essay traces the debate's history in computer science and calls for a bottom-up, market-driven approach to steering AI development.

  7. Every40

    How Codex Happened: OpenAI's Comeback in AI Coding

    OpenAI's Codex, once a scrappy cloud effort that failed to find product-market fit, became the centerpiece of the company's AI coding comeback after pivoting to a terminal and code-editor version. In summer 2025 testing, GPT-5 inside Codex felt fast but overly cautious compared with Anthropic's Claude Code, which let users delegate entire tasks to Opus 4 and walk away.

  8. Meta AI Research Blog88

    Meta releases Muse Spark Open, an 816B open-weight multimodal model that runs on two Mac Studios

    Meta released the weights for Muse Spark Open, an open-weight natively multimodal model with 816B total parameters (42B active) and a 1M-token context window, positioned for coding and agentic work. It is the first model in its class to run on a desktop using two Apple M5 Ultra Mac Studios with the MLX framework, and it is available through API partners including Baseten, Fireworks AI, Groq, Modal, Nebius, OpenRouter and Together AI, with weights on Hugging Face, Amazon SageMaker AI, Google Cloud Model Garden and Microsoft Foundry Models. Meta says the model ships with auto-tunable Helion GPU kernels in SGLang, delivering 1.03-3.95x kernel-level speedups on NVIDIA and AMD GPUs and a 3% end-to-end speedup on AMD MI350.

    Why it matters: Meta releases open weights for an 816B-parameter multimodal model that runs on two Mac Studios, giving developers a local option for long-horizon coding and agentic tasks.

  9. Apple Newsroom60

    Apple launches iPhone Duo, its first foldable iPhone with a 7.58-inch inner display

    Apple announced iPhone Duo, its first foldable iPhone, featuring the largest iPhone display yet: a 7.58-inch inner screen that is 50 percent larger than iPhone 18 Pro Max's display, plus a 5.4-inch outer display covering 90 percent of iPhone 18 Pro's screen area. It runs on the A20 Pro chip with a dual-battery architecture, and iOS 27 is reworked for the foldable form factor, including Split View for showing Siri and Mail side by side. Siri AI, powered by Apple Intelligence, is rolling out in beta with iOS 27 in English first and will not initially be available in the EU; Apple notes Siri AI and other new Apple Intelligence features will not be available in China.

    Why it matters: Apple's first foldable iPhone pairs a 7.58-inch inner display with a 5.4-inch outer screen and an A20 Pro chip, showing how iOS 27 and Siri AI are being adapted to a foldable form factor.

  10. Goodfire Research62

    Goodfire research essay argues neural networks are already clean and superposition was misread

    Goodfire researcher argues that the canonical evidence for polysemanticity and superposition in neural networks was misread, and that what looked like arbitrary packing is mostly a motif he calls recursive generalization. Retraining InceptionV1 with ordinary architectural improvements and no change to model size or training data made 99-100% of neurons in its last layer easily interpretable, and preliminary audits of Qwen3-32B estimate roughly 90% of neuron poles have one main job. The essay concludes networks use one learning principle rather than two, so interpretability can proceed by studying neurons and circuits directly instead of first training decomposition models such as sparse autoencoders.

    Why it matters: A first-hand reinterpretation of the field's canonical polysemanticity evidence, arguing that networks generalize recursively rather than pack unrelated concepts, which would make direct mechanistic study viable.

  11. Modal Blog78

    Meta's Muse Spark Open weights released and available on Modal Auto Endpoints

    Meta released the weights for Muse Spark Open, and Modal now serves the model through its Auto Endpoints with per-token billing. The model targets agentic coding and long-horizon tasks, with native multimodal understanding, a 1M-token context window, multi-step planning, sequential tool use and failure recovery. Meta reports 76.4 on Terminal-Bench 2.1 with the Terminus 2 agent and 60 on DeepSWE 1.1 using its preferred harness at xhigh reasoning effort.

    Why it matters: Meta's open-weight Muse Spark Open is now served on Modal, giving teams a per-token way to try an agentic coding model with a 1M-token context window.

October 14Wednesday
  1. GitHub Blog · AI & ML38

    GitHub's secret detection model: finding more secrets with less noise

    GitHub trained a fine-tuned ModernBERT classifier to replace its previous LLM-based pipeline for generic secret detection, improving owner-weighted alert precision by 11.5% and cutting median post-push scan execution time by 86%. The two-stage pipeline uses regex candidate generation followed by context-aware classification, which also lowered serving cost by more than 98% versus remote LLM calls. The efficiency gain made it practical to extend generic secret detection to push protection, returning results before changes are pushed.

  2. Ed Zitron · Where's Your Ed At58

    Ed Zitron: How Annualized Run Rates and Hidden Startup Economics Inflate the AI Bubble

    Ed Zitron argues that annualized run rates used by OpenAI, Anthropic and AI startups are undefined, non-comparable figures that obscure real quarterly revenue. He cites OpenAI's reported $70 billion ARR later revised to $50 billion, Anthropic's $65 billion run rate, and Harvey's gross margin falling from about 50% to -50% by June, contending that AI startups resell tokens at a loss and that OpenAI and Anthropic growth depends on continued venture funding.

  3. Apple Newsroom12

    Les Benjamins' Bünyamin Aydın Debuts "From Earth to Sky" Collection Made with iPad Pro

    Les Benjamins designer Bünyamin Aydın unveiled his "From Earth to Sky" collection, created with iPad Pro and Apple Pencil Pro and filmed on iPhone 18 Pro Max in Cappadocia's Rose Valley. The show marks the Istanbul brand's 15th anniversary with 10 looks inspired by three native flowers and three custom hot-air balloons. Aydın used Freeform, Procreate, and iPad Pro's M5 chip across the design workflow.

  4. Modal Blog71

    Modal crosses $1B annualized run rate and raises $750M Series D at $15.75B valuation

    Modal crossed $1B in annualized run rate and raised a $750M Series D led by Accel at a $15.75B post-money valuation, with all existing investors participating. The company says usage has tripled since its Series C five months ago and grown more than 15x year-over-year, and that its Sandboxes product alone is at a $300M run rate with close to 3 billion sandboxes spun up. Modal also describes rebuilding its own file system, container runtime, scheduler and networking on a capacity pool spanning 20+ clouds, and points to recent launches in multi-node training, Modal servers and generally available VMs for Sandboxes.

    Why it matters: Modal's post explains why AI workloads pushed it to rebuild its own file system, container runtime and scheduler across 20+ clouds, a concrete view of how the AI infrastructure stack is diverging from the cloud stack.

October 13Tuesday
  1. Recode China AI34

    DeepSeek Engineer's Viral Blog on AI Risk and Anthropic Draws Global Media Attention

    A blog by DeepSeek machine learning and kernel engineer Liu Shengyu went viral after he questioned whether frontier AI can now optimize its own low-level GPU code and compared Anthropic CEO Dario Amodei to Adolf Hitler. The post drew coverage from outlets including South China Morning Post and Business Insider, amplifying debate over who should control advanced AI.

  2. LiveKit Blog18

    Troubleshooting LiveKit Egress: Common Errors and Fixes

    LiveKit published a troubleshooting guide for Egress, covering common errors such as the 412 "start signal not received" from templates that never call START_RECORDING, 404 participant or track not found within 30 seconds, and 400 upload failures from invalid S3/GCP/Azure credentials. It also addresses layout, webhook, and self-hosted Egress issues, and notes the guidance applies to both the current StartEgress API and the deprecated source-specific APIs.

  3. Modal Blog58

    Modal tests mixed-precision, mixed-accelerator RL samplers on Qwen3.6-35B-A3B

    Modal trained Qwen3.6-35B-A3B on agentic coding with samplers spread across six accelerator types (B200, B300, H100, H200, A100, RTX PRO 6000) at NVFP4, FP8 and BF16 precision, feeding one BF16 trainer on B200. Plain GRPO collapsed on both mixed and homogeneous samplers, but adding masked importance sampling plus score centering, with or without top-p mask replay, trained stably to the end and reached SWE-bench Pro V2 pass@4 on par with homogeneous samplers, at roughly 15% lower sampler cost per update. The mixed samplers ran with an average sampler-trainer KL 4.7x that of the homogeneous setup, and the authors argue stability depends on keeping the mismatch bounded rather than low.

  4. GitHub Blog · AI & ML36

    GitHub Copilot for Beginners: Four Ways to Teach Copilot About Your Codebase

    GitHub's Copilot for Beginners series explains four ways to give Copilot project context: custom instructions in `.github/copilot-instructions.md` plus path-specific files in `.github/instructions`, agent skills in `.github/skills`, custom agents in `.github/agents`, and MCP servers that connect Copilot to external tools such as Playwright. The post walks through each mechanism and how they combine, and points readers to the community Awesome Copilot collection for ready-made examples.

  5. Amazon Science75

    Amazon Web Services researchers propose priming to convert pretrained Transformers into hybrid state space models without full retraining

    Amazon Web Services researchers introduce priming, a method that reuses pretrained Transformer weights to cut the pretraining cost of hybrid state space architectures by more than 99%, using under 0.5% of the source model's pretraining token budget. In controlled comparisons, Gated KalmaNet outperformed Gated DeltaNet and Mamba-2, and a 32B GKA-based hybrid matched its source Transformer on reasoning benchmarks while decoding up to twice as quickly at 128,000 tokens of context. The team released 8B and 32B primed hybrid models, training code, Triton kernels, and a vLLM serving plugin under the Apache 2.0 License.

    Why it matters: Amazon Web Services researchers show that reusing pretrained Transformer weights can cut hybrid architecture pretraining cost by over 99%, making controlled comparisons of state space layers practical.

  6. Apple Newsroom58

    Apple introduces HomeView, a $299 smart home display with Siri AI

    Apple introduced Apple HomeView, a 7-inch Liquid Retina smart home display that combines entertainment, FaceTime calls, home monitoring and hands-free automation, priced at $299 in the U.S. The device recognizes users by face to surface their personal content and runs Siri AI, a new more conversational version of Siri that can draw on personal context when paired with an Apple Intelligence-enabled iPhone or iPad. Pre-orders opened October 13 in Australia, Canada, India, New Zealand, the UK and the U.S., with availability starting Friday, November 13; Apple says HomeView is not for sale in the European Union because Siri AI is currently unavailable in iOS there.

  7. Every77

    Google DeepMind releases Gemini 4 Argon, returning Google to the frontier

    Google DeepMind released Gemini 4 Argon today, marking Google's first major flagship model upgrade since Gemini 3 Pro last November and a minor 3.1 Pro bump in February. The release follows a summer in which Google shipped four rapid updates to its faster, lower-cost Flash line while the promised Gemini 3.5 Pro stayed in testing, and Every frames Argon as putting Google back in the frontier conversation.

    Why it matters: Google DeepMind releases Gemini 4 Argon, its first flagship upgrade since Gemini 3 Pro, returning Google to the frontier model conversation.

  8. Apple Newsroom58

    Apple unveils new Apple TV 4K with Siri AI and hands-free far-field microphones

    Apple announced the fourth-generation Apple TV 4K, featuring Siri AI, a new version of Siri powered by the next generation of Apple Intelligence that offers a full-screen conversational search experience. Built-in far-field microphones enable hands-free Siri use, and with an Apple Intelligence-enabled iPhone the assistant can answer personal-context requests such as playing a show a friend recommended. The device is priced at $299 in the U.S., with pre-orders open in the U.S. and 29 other countries and regions and in-store availability starting Friday, November 13; Siri AI is available as a beta on Apple TV 4K (2nd generation or later) in eight English-language markets at launch.

October 12Monday
  1. Import AI38

    Import AI 476: Creativity and RSI, Worse-Than-China Open Weight Models, and AI in Space

    Import AI 476 covers a new benchmark showing frontier LLMs still trail humans on intuitive visual reasoning, with GPT 6 Astra scoring 53.6 pass@1 versus a 93.1 human baseline. It also discusses AI systems improving at "research taste" via the TasteVal benchmark, and argues AI could accelerate space industrialization and create compounding first-mover advantages.

  3. Meta AI Research Blog82

    Meta releases Muse Wonder and Muse Wonder Full frontier models for personal and professional agents

    Meta introduced the Muse Wonder series, comprising Muse Wonder and Muse Wonder Full, which it says delivers state-of-the-art performance on personal and professional agentic tasks, memory and coding. Muse Wonder is available today through the Meta Model API and Muse Code, while Muse Wonder Full is available today for Muse Code subscribers. The models pair Active Memory, which carries context across conversations and retrieves task-relevant details, with Active Focus, which sustains work on long-running goals and checks in before important actions; Meta says both were trained at its gigawatt-scale Prometheus data center.

    Why it matters: Meta's Muse Wonder series claims state-of-the-art results on agentic, memory and coding benchmarks, with Active Memory and Active Focus as the two capabilities it is built around.