Updated
Updated
Showing low-relevance items too. Hide low-relevance items
LM Studio@lmstudioAI score36AILM Studio introduced Auto Review, which auto-approves shell tool requests using AST parsing, command matching, and a reviewer subagent. In the team's internal use, about 82% of commands were approved before reaching LLM review.
Meituan LongCat@Meituan_LongCatAI score62AIMeituan LongCat evaluated 7 frontier models on 36 AI R&D tasks covering 756 trajectories, looking beyond final scores. Of 252 solutions, only 3 were novel approaches, and most adapted or combined established techniques. The authors conclude that current agents work more like engineering optimizers than autonomous researchers, with reliability, experience reuse, and novelty still open challenges.
RadixArk@radixarkAI score34AIRadixArk introduced LoRA SFT in Miles-diffusion for fast, targeted post-training of diffusion models. The company trained a rank-64 LoRA adapter for MiniMax H3 to improve physical realism, using 254 curated training windows and under 3 hours on 8 GPUs. The adapter can be exported to safetensors and served directly with SGLang without retraining the full model.
AIAnthropic is introducing the Model Hardware Standard (MHS), a new standard for AI agents to safely operate physical equipment in scientific research and advanced manufacturing. MHS began as part of a beneficial deployments project with HHMI Janelia Research Campus and is evolving into a wider industry effort. It is now in research preview with select partners.
Noah Zweben@noahzwebenAI score36AIAnthropic says on-call is the most popular use case for Claude Tag, which it uses internally to handle alerts. The linked post explains how to set it up so Claude can resolve issues that would otherwise wake engineers at 3am.
AIAugment Code introduces Cosmos Advisor, an expert that can answer product questions, configure agents, and deploy automations from a single conversation. The company says a company-specific agent can be set up in about ten minutes, without a handoff to an implementation team. Advisor draws on the current Cosmos knowledgebase and reusable expert designs, such as incident response, and it works within Object-Level Access Control.
AIAugment Code's two-engineer Cosmos Advisor team built a Feedback Triager agent to handle product feedback that grew to about 30 threads per week, which had consumed an estimated 90% of team time. The agent investigates each Slack report through root-cause analysis, answers questions, routes issues to other teams, files tickets, and hands clear fixes to a PR Author agent. Humans retain prioritization and product decisions.
Waymo@WaymoAI score34AIWaymo introduced the Ojai, a new vehicle that will run the same Waymo Driver autonomous system. The post gives no further details on specifications, deployment timing, or availability.
AIAnthropic is building the Model Hardware Standard (MHS), a common way for AI models to connect to lab and manufacturing equipment and operate it with safety limits built into each device. MHS started as a collaboration between Anthropic and HHMI Janelia Research Campus and is launching as a research preview with partners across science, robotics, and manufacturing.
Why it matters: The source describes a standard for connecting AI models to lab and manufacturing hardware, which matters for anyone building automated experimentation workflows.
Ali Ghodsi@alighodsiAI score22AIAli Ghodsi recommends branching a database to guard against AI agents permanently wiping data, citing Neon Lakebase and the command `neonctl branches create --name newbranch`. The suggestion follows a quoted report in which Claude ran `rm -rf` on a developer's home directory while testing a sandbox, deleting everything.
Tinker@tinkerapiAI score41AIalphaXiv is turning research papers from static artifacts into live research that grows and branches, with agents running their own experiments. Tinker says it makes running these experiments easy for both agents and people. Via alphaXiv's background post, its autoresearch tool lets Claude or Codex agents replicate and experiment on any arXiv paper, with agents launching concurrent RL runs through Tinker for post-training.
AIAnthropic and HHMI Janelia Research Campus developed the Model Hardware Standard (MHS), a standard for AI agents to safely operate physical equipment in scientific research and advanced manufacturing. MHS is now in research preview with select partners, and the video describes how it was developed and how it can accelerate research.
Replicate@replicateAI score10AIReplicate has made Google's Gemini Omni 1.1 available to try through a hosted model page. The post provides only a link to the model on replicate.com, with no specifications, benchmarks, or pricing mentioned.
Replicate@replicateAI score62AIReplicate announced that Gemini Omni 1.1 Flash from Google DeepMind is now live on its platform. The update adds scene extension, control of a shot's starting and ending frames, video input references, upscaling to 4K, and 360p fast prototyping.
LMSYS Org@lmsysorgAI score20AILMSYS Org posted a link to a blog post about MiniMax H3 on H200 hardware, dated August 27, 2026. The post itself offers no benchmark figures, specifications, or other details beyond the link.
LMSYS Org@lmsysorgAI score47AIMiniMax-H3 on 8×H200 GPUs reaches 1.85–1.95x lossless speedup over Diffusers without approximation, with fixed prompts, seeds, resolution, FPS, and 50 denoising steps. Adding step reuse and sparse attention raises speedup to as much as 6.24x, but quality varies by workload, with SSIM from 0.76 to 0.91. Two presets trade off the two: a conservative Cache-DiT setting gives 2.99x at 0.90–0.98 SSIM, while a faster SubBlock 0.75 plus Cache-DiT stride gives 4.90–5.93x at 0.77–0.92.
Vercel@vercelAI score40AIVercel has open-sourced vgpu.sh, a minimal, agent-first WebGPU library for shipping shaders on Vercel. It can run in the browser or headless in Node.js, render in CPU sandboxes and CI tests, and create reusable .wgsl modules.
Unsloth AI@UnslothAIPickAI score70AIUnsloth says GLM-5.3-Flash can run locally, with a 3-bit GGUF version needing 128GB of RAM and the 1-bit version working on 102GB of RAM or VRAM. The guide's table lists memory needs from 100GB at 1-bit to 650GB at BF16, and reports that the 1-bit quant keeps 71% of top-1% accuracy while being 85% smaller than BF16.
Why it matters: The guide gives concrete memory requirements for each quantization level, which helps readers judge whether the model fits their hardware.
Leandro von Werra@lvwerraAI score22AIPollen Robotics has unveiled Microduck, a 25 cm open-source biped with 15 actuators and sensors including a camera, speaker, and LiDAR that users can train with reinforcement learning. The robot ships with more than half a dozen pre-trained policies for walking, sitting, roller-skating, and picking up objects with its articulated beak, and costs less than $400.
Bryan Catanzaro@ctnzrAI score46AINVIDIA's DLSS 4.5 Ray Reconstruction is now available, using a second-generation joint denoiser and super-resolution model. According to the post, it delivers much better image quality at the same compute cost, pushing the trade-off between image quality and rendering cost further.
AICursor Cloud Agents no longer require a connected GitHub or other third-party SCM provider to begin work. Users select "Start from scratch" in the repo picker, and Cursor creates an Origin repo in the background that can be saved as a private or internal repo via "Create repo." Cursor also now port-forwards the cloud agent's live environment to the browser for previews, and a connected Vercel account lets users publish a live URL.
AIIn an interview with Jiazi Guangnian, Renmin University information school dean Chai Yunpeng describes his team's social simulator, which runs over 13.5 million AI agents calibrated against the CGSS survey data. He argues that social world models are the missing piece for AI agents that must interact with people, and that the startup Jingtong Technology has raised two funding rounds in two months.
AIThe Google Developers Blog post describes how deep learning can analyze the large, image-like sensor data from astroparticle observatories such as the Pierre Auger Observatory and IceCube. The author argues these methods could improve instrument sensitivity and reveal patterns in cosmic-ray and neutrino signals that traditional analysis techniques miss.
Bryan Catanzaro@ctnzrAI score62AINVIDIA and AWS are expanding their partnership across GPUs, CPUs, networking, open models and software. The announcement cites 2 million additional NVIDIA GPUs across AWS infrastructure, the NVIDIA Vera CPU coming to AWS for agentic AI, NVLink Fusion with NVHBM memory, and 100,000 GPUs for U.S. government AI factories on secure AWS infrastructure.
Replicate@replicateAI score32AIReplicate has made Wan 3.0 Prime, the accelerated variant of Alibaba's Wan 3.0, available for text-to-video, image-to-video, and reference-driven workflows. The model generates clips up to 30 seconds long in a single shot with integrated audio-visual generation.
AIReplit has made Intelligent Model Routing available to all users, automatically matching each task with a model based on quality, speed, and cost. In the company's testing, the feature delivered the same output quality at 65% lower cost than the previous version of Max Mode. Enterprise administrators can restrict routing to a company-approved set of models.
LM Studio@lmstudioAI score57AILM Studio announced that Z.ai's GLM-5.3-Flash, previously previewed as Ox Alpha, is available in LM Studio Bionic. The source says the model outperforms GLM-5.2 at 9-10x lower cost, supports image input, and is served from US-based servers with ZDR enabled by default.
LM Studio@lmstudioAI score18AILM Studio published a model page for GLM-5.3 Flash on its website. The post contains only a link to the model page and no further details about the model's specifications or capabilities.
Matei Zaharia@matei_zahariaAI score38AIMatei Zaharia said an AI agent's `rm -rf` on a developer's home directory shows why cloud storage and databases must be branchable and recoverable. He pointed to Neon and Databricks as building that capability, arguing agentic development will need a new kind of cloud infrastructure.
Michael Truell@mntruellAI score60AIGrok Bot is now available to all standard Grok and Cursor subscribers, with SuperGrok and Cursor Pro subscribers included. Cursor's Michael Truell says users are delegating tasks ranging from running small e-commerce businesses to testing production software. Weekly usage limits are also being reset for all users.
Sundar Pichai@sundarpichaiAI score42AIGoogle has released Gemini 3.5 Transcribe, a speech-to-text model that auto-detects over 85 languages and handles multiple speakers. It also supports custom vocabulary adaptation for specialized jargon. The API is available now in Google AI Studio and Gemini Enterprise.
SpaceXAI@SpaceXAIAI score42AILiveKit announces that developers can build voice agents using Grok Voice models, with full ZDR support. LiveKit's example patient intake agent cascades Grok STT, Grok 4.3, and Grok TTS through LiveKit Inference in a single AgentSession, with no separate API key or billing.
LMSYS Org@lmsysorgPickAI score65AIZ.ai released GLM-5.3-Flash, a 320B-A18B model, with day-0 support in SGLang, after appearing earlier as ox-alpha. The post calls it the first native multimodal model in the GLM-5 series and says it outperforms GLM-5.2 at one-tenth the cost, with stable 1M-token long-context performance.
Why it matters: The post reports GLM-5.3-Flash's native multimodal design, its efficiency claims, and day-0 SGLang support, which bear on running it in practice.
Replicate@replicateAI score16AIReplicate is letting users try four Recraft V4 variants: styles, styles SVG, styles pro, and styles pro SVG. The post provides links to each model on Replicate, with no details on pricing, benchmarks, or capabilities.
LMSYS Org@lmsysorgAI score60AISGLang announced Day-0 support for Qwen3.8-Flash-Next, a 125B MoE model with 6B active parameters and 51B N-gram embeddings, in collaboration with Alibaba Qwen, NVIDIA, and AMD. The post reports 540 tok/s decode speed at BS=1 on NVIDIA B200 (TP4) with an NVFP4 checkpoint, and says N-gram host offloading saves 23.5 GiB VRAM per GPU and raises KV capacity by 78.5%.
AIMicrosoft's Accelerating Frontier Transformation series says AI is helping organizations deliver more personalized engagement at scale and give staff time back for relationships. Examples include Lifeline Australia using AI for service insight, Brisbane Catholic Education personalizing curriculum for students with Copilot, and Uniting NSW.ACT's Buddy platform cutting some frontline tasks from 10 to 15 minutes to one to two.
Z.ai@Zai_orgAI score29AIZ.ai has published standard API pricing for GLM-5.3-Flash, charging $0.15 per 1M input tokens and $0.50 per 1M output tokens. Cached input tokens are priced at $0.03 per 1M tokens.
AIDeepSeek V4 Pro 0813 scored 95.2% on SWE-Bench Verified, ahead of Kimi K3 at 92.6% and Fable 5 at 85.4%, in Fireworks AI's eval runs. It costs $0.309 per solved task on SWE-bench versus $0.808 for Fable 5, and it is available through Fireworks serverless and dedicated endpoints, with SFT, DPO, and RFT training support. Its 1M-token context window and native tool calling target long-horizon agentic workloads, though its Java accuracy on Aider Polyglot (48.9%) trails Fable 5 (74.5%).
AIDeepSeek V4 Pro 0813 recorded zero refusals across 840 adversarial security tasks in CyberGym testing, solving them at about half the cost per success of the top-scoring model tested, Kimi K3. In the 697-task common cohort, V4 Pro reached a 53.7% reward rate at $2.50 per solved task, versus 47.6% and $9.64 for GPT-5.5 and 5.9% and $33.28 for Claude Opus 4.8.
AIGoogle Cloud has added native TPU support to vLLM and engineered optimizations to serve the Qwen3-Embedding-8B model on Cloud TPU, targeting 4K+ token text and 15K+ token multimodal inputs. The work addresses tensor alignment, lazy-loading, compilation pre-warming, and long-context pooling, with a cosine similarity pass threshold of at least 0.999 for text and 0.995 for multimodal inputs against XPU reference vectors.