Updated
Updated
Showing low-relevance items too. Hide low-relevance items
Joshua Achiam@jachiam0XAI score4AIJoshua Achiam says he took a quiz or test a second time after the first attempt returned a Dean Ball easter egg. He calls the result very reasonable and says landing dead center on the map feels on-brand.
Arena.ai@arenaOfficialAI score17AIArena says it is proud to work with Khosla Ventures, which describes Arena as having built the industry's most important AI leaderboard. Khosla Ventures says Arena is now measuring how AI agents behave in real-world settings.
AIMeta says its closed-loop liquid cooling recirculates water in a sealed system, so its data centers use less water annually than an average US golf course. The company also says it pays for the new generation and transmission its facilities require, including in Louisiana under its Entergy agreement, and that data centers create construction and operations jobs.
elvis@omarsar0XAI score46AIRSIGym provides a research agent with training, inference, evals, and sandboxes as callable services, so it spends its budget on experiments rather than rebuilding infrastructure. With Opus 5 as the researcher, the improved system rose from 17.67% to 50.33% on SWE-bench Verified. The post also highlights a way to measure co-evolution between harnesses and models.
Modal@modalOfficialAI score16AIModal says AI for science requires rapidly changing models and fluctuating GPU needs, including folding models, protein diffusion, single-cell LLMs, and agent swarms. The post promotes computational biology leaders from GSK, Anthropic, and Achira speaking on the main stage at Runtime.
Odyssey@odysseymlOfficialAI score31AIOdyssey has released Odyssey-3, which it describes as a major leap toward world models that power physical AI, generate training environments, and enable new human experiences. The post invites readers to try Odyssey-3 at the company's website but gives no specific benchmarks, parameter counts, or pricing.
Odyssey@odysseymlOfficialAI score34AIOdyssey says its Odyssey-3 model's learned world knowledge can be adapted by physical AI developers to control robots, power humanoids, drive cars, and fly drones. The post describes this as a capability for autonomous machines generally, without providing benchmarks, specifications, or availability details.
Odyssey@odysseymlOfficialAI score22AIOdyssey announced Odyssey-3, which it says can generate diverse interactive environments limited only by the user's imagination. The post provides no further technical details, benchmarks, or availability information.
Odyssey@odysseymlOfficialAI score38AIOdyssey announced Odyssey-3, a foundation world model it says enables applications in physical AI, human experiences, and training intelligences. The company highlights agents learning from experience inside Odyssey-3 while working toward objectives.
Odyssey@odysseymlOfficialAI score22AIOdyssey-3 Pro achieved a score of 66.1 on Physics-IQ Verified's video-to-video benchmark, the highest reported score so far. Physics-IQ evaluates physical behavior across fluid dynamics, optics, solid mechanics, magnetism, and thermodynamics.
Odyssey@odysseymlOfficialAI score40AIOdyssey has launched Odyssey-3, which it describes as its most powerful foundation world model yet. The company says the model sets a new state of the art on Physics-IQ and can power robots, train AIs, and generate interactive experiences, and it is available to try for free today.
Odyssey@odysseymlOfficialAI score42AIOdyssey-3 is a new world simulator that generates interactive environments from a prompt in real time. It learns representations of physics, dynamics, and cause-and-effect from a broad dataset of visual observations.
LlamaIndex 🦙@llama_indexOfficialAI score8AILlamaIndex says Markdown is its default output for document parsing because it preserves headings, lists, and tables, which helps models read content correctly. The post notes that parsers can extract every word yet lose which column a number belongs to, forcing models to guess. For tables with merged headers, LlamaIndex switches to HTML.
Aravind Srinivas@AravSrinivasXAI score22AIPerplexity's Decider V1.1 ranked first on DecisionBench while also having the lowest cost, according to a post highlighting the result. The benchmark results cited include 949 shared text cases, 93.9% accuracy, a 534 ms median latency, and $0.016 per 1k decisions.
Stanford HAI@StanfordHAIOfficialAI score22AIStanford HAI associate directors Risa Wechsler and Russ Altman told incoming Stanford students, faculty, and staff that AI agents can help researchers write code and tackle more ambitious questions. They stressed that AI-generated results need rigorous, reproducible methods, measured uncertainty, and careful attention to missing data, systematic errors, and biased models. Altman also argued that labs should preserve mentorship and interdisciplinary collaboration while adopting AI tools.
AIJevman is an open-source Pac-Man benchmark where AI models play 100 games each against the classic scripted ghosts. Each model gets a maze state at every junction and returns a direction probability, with answers over 2 seconds replaced by a backup rule. Community models can join the leaderboard by submitting games that CI replays to verify their scores.
lauren@potetoXAI score25AIShopify has launched new connectors that let merchants connect Grok to ask questions about their business. The connectors also allow Bot to build a team of agents that help run the business.
Fireworks@FireworksAI_HQOfficialAI score34AIZeta Global unveiled the Athena Inference Model (AIM), being developed with Fireworks using Nvidia Nemotron, as part of its AthenaOS enterprise intelligence announcements. The company also said it is doing early work with partners on a specialized chip through a capital-light approach.
Amazon Science@AmazonScienceOfficialAI score5AIAmazon is hosting a live AI rap battle demo at its booth at the COLM conference, where attendees can try to out-rhyme an AI agent. The booth also features talks on multi-agent systems and responsible AI for science.
LangChain@LangChainOfficialAI score34AILangChain has built Restock, an office supply agent that works inside Slack and can find real products, prepare purchases, and pay for them. A person approves each order, which is reviewed in Slack and approved through Stripe's Link agent wallet, built on MPP and Managed Deep Agents.
Goodfire@GoodfireAIOfficialAI score22AIGoodfire has published research on deploying cyber monitors in production on Kimi K3. The post links to the full research write-up and invites organizations to get in touch about deploying these monitors.
Goodfire@GoodfireAIOfficialAI score21AIGoodfire reports that running its probes during model inference maintains the same throughput with no added latency. The company attributes this to infrastructure engineering, including kernel-level optimizations and a custom inference server.
Goodfire@GoodfireAIOfficialAI score25AIGoodfire says a static battery of attacks from FAR AI red-teamed its monitor, and the monitors reduced successful universal jailbreaks to 0. The same testing cut total jailbroken interactions by 97%.
Goodfire@GoodfireAIOfficialAI score22AIGoodfire says its monitor Pareto dominates an LLM judge on recall and false positive rate while using 50x less compute. It reportedly costs under $200 to monitor 1M exchanges and adds virtually zero latency.
Goodfire@GoodfireAIOfficialAI score28AIGoodfire built cybersecurity monitors for Kimi K3 and GLM 5.3 that it says are more accurate, 50x faster, and 50x cheaper than an optimized LLM judge. External red-teaming by Far AI Research reportedly found the monitor greatly reduces universal jailbreaks.
Daniel Han@danielhanchenXAI score38AIUnsloth now supports OS-level sandboxing on Linux via bwrap, on Mac via seatbelt, and on Windows via Microsoft's MXC. Per-tool-call latency is under 100ms across all three, and its software-style sandboxing with regex AST checks adds about 3ms. The Windows integration was built in collaboration with Microsoft.
AIPeriodic Labs' Liam Fedus and Ekin Dogus Cubuk explain why scientific discovery differs from math and coding, and why experiments remain the ground truth. They describe reinforcement learning grounded in physical experiments, AI-driven materials characterization, and the view that failed experiments can be valuable training data. The transcript was truncated before the discussion of giving lab instruments "140 IQ" was completed.
Suno@sunoOfficialAI score22AISuno announced that Albums are now live, letting users combine songs into a full release, set artwork, arrange the tracklist, and publish when ready. Existing playlists can be converted into Albums without rebuilding them from scratch.
Google Research@GoogleResearchOfficialAI score14AIGoogle Research is presenting EnvHarness, a flexible framework that enables co-evolution between LLM agents and their training environments, at the #COLM2026 Google booth #107 today at 11:00 AM PT. The post notes that static environments limit agent growth, and EnvHarness is described as a plug-in architecture that dynamically reshapes environment behaviors to improve reinforcement learning and adaptability.
OpenClaw🦞@openclawOfficialAI score34AIOpenClaw released its v2026.9.9 patch, adding support for GPT-6.1 Sol in Codex and Claude Haiku 5.5. The update also improves recovery from failed updates, fixes missing iMessage replies, and resolves scheduled-job issues. The release credits 90 contributors.
Google Gemma@googlegemmaOfficialAI score27AIGoogle Gemma has published a developer guide for EmbeddingGemma 2, with code snippets to help developers start searching beyond text. The post directs readers to the full guide on the Google Developers Blog.
Google Gemma@googlegemmaOfficialAI score44AIGoogle Gemma announces a developer guide showing how to embed text, code, images, video, audio, and interleaved inputs with EmbeddingGemma 2 using the sentence-transformers library. The guide outlines a four-step workflow: loading the model, embedding text and code with task prompts, embedding multimodal inputs, and optionally truncating dimensions with Matryoshka.
lauren@potetoXAI score20AILauren Tan (@poteto) asks a bot to monitor interesting ways people use Grok Bot on X and send a daily 9am Slack digest. The request also asks for the coolest relevant ideas plus suggestions for improving her workflow.
elvis@omarsar0XAI score34AIVoice Arena's Monsoon ASR dataset fine-tuned Whisper Medium on Bengali FLEURS, reducing LLM word error rate from 85.27% to 7.65%. The corpus spans 100,000 hours across 50 languages, and Voice Arena says more than 80 organisations have asked to license it since its launch a week ago.
Nous Research@NousResearchOfficialAI score27AINous Research says Hermes Agent is now live and featured in the Microsoft Store, allowing one-click installation. The post also says more Windows ecosystem updates are coming soon.
AIAWS published a reference architecture for running multiple teams on one Amazon SageMaker HyperPod EKS cluster, with each team isolated in its own Kubernetes namespace. The design combines AWS IAM Identity Center for authentication, per-team SageMaker AI domains, HyperPod Task Governance for fair resource allocation, and namespace-level cost allocation for per-team spend visibility.
jason@jxnlcoXAI score8AIJason Liu says he lacked the vocabulary, not the ambition, to design objects in CAD for 3D printing, so he learned printing, part efficiency, and shape terminology. He argues AI is like magic that still requires learning the right spells and incantations, and that it has made him want to understand the subject more deeply.
AIJetBrains has released Mellum2.1, a 12B mixture-of-experts thinking model with 2.5B active parameters, under Apache 2.0 on Hugging Face. Post-training reinforcement learning in real software repositories raised SWE-bench Verified from 2.0 to 47.0, according to JetBrains' self-reported results. Qwen3.5-9B still leads on SWE-bench Pro, GPQA Diamond and AIME, and GGUF builds start at 7.0 GB for local use.
elvis@omarsar0XAI score22AINatura AI's Interface is a ring that lets users press and hold to speak requests to AI agents such as Claude Code, Codex, or Hermes, then release to send them. The post argues that screenless interfaces may define the next phase of agent use, since handing work to agents is currently slowed by pulling out a phone. Early-adopter pricing is $99, with shipping slated for January.
AIJabil senior director Thomas Brown says humanoid robots are entering a phase of tens of thousands of units, where manufacturability, cost structure, and quality become central. He says Jabil works with developers to cut costs for scale, while compute and memory prices remain a pain point, and that humanoids make sense in factories and warehouses while mobile arms still suit high-speed tasks.