# AlphaSignal Full Index

Site map: https://alphasignal.ai/llms.txt

## Recent articles

- Title: Artificial Analysis Reshuffles Its Intelligence Index to Separate Frontier Models
  URL: https://alphasignal.ai/news/artificial-analysis-reshuffles-its-intelligence-index-to-separate-frontier
  Summary: Artificial Analysis rolls out Intelligence Index v4.2 early, swapping GPQA Diamond for long-horizon agentic work and multi-thousand-page document reasoning.
  Published: 2026-09-06T00:28:27.000Z
  Topics: benchmarks, reasoning
  Company: Artificial Analysis

- Title: Meta's AIRA₃ Beats 4,000 Human Teams to Win Kaggle Gold
  URL: https://alphasignal.ai/news/meta-s-aira3-beats-4-000-human-teams-to-win-kaggle-gold
  Summary: Meta's next-generation autonomous research agent AIRA₃ placed 8th of roughly 4,000 human teams in an NVIDIA-run Kaggle contest to fine-tune Nemotron.
  Published: 2026-09-05T16:17:37.000Z
  Topics: agents, reasoning
  Company: AI at Meta

- Title: NVIDIA's Nemotron Beats the Best Human at the Coding Olympics
  URL: https://alphasignal.ai/news/nvidia-s-nemotron-beats-the-best-human-at-the-coding-olympics
  Summary: NVIDIA's fine-tuned Nemotron system scored 535.4 out of 600 at IOI 2026, topping the best human contestant under identical contest conditions.
  Published: 2026-09-05T00:28:00.000Z
  Topics: post training, benchmarks
  Company: NVIDIA AI

- Title: GitHub Restores Star Growth Tracking Without Exposing Who Starred
  URL: https://alphasignal.ai/news/github-restores-star-growth-tracking-without-exposing-who-starred
  Summary: GitHub added a REST endpoint that returns weekly star counts with timestamps, restoring star-history tracking that broke when stargazer listings were locked down.
  Published: 2026-09-05T00:04:30.000Z
  Topics: api, development
  Company: GitHub

- Title: Artificial Analysis Rebuilds Its AI Leaderboard to Stop Models From Gaming It
  URL: https://alphasignal.ai/news/artificial-analysis-rebuilds-its-ai-leaderboard-to-stop-models-from-gaming-it
  Summary: Artificial Analysis pushes Intelligence Index v4.2 with private test sets, a 4,592-page PDF reasoning benchmark, and a new agentic knowledge-work suite.
  Published: 2026-09-04T22:26:29.000Z
  Topics: benchmarks, llms
  Company: Artificial Analysis

- Title: Perplexity's ROSE Serving Stack Beats vLLM on Speed and Latency
  URL: https://alphasignal.ai/news/perplexity-s-rose-serving-stack-beats-vllm-on-speed-and-latency
  Summary: Perplexity open-sources details of Ivy, Tulip, and ROSE, its Rust plus Python embedding stack that beats vLLM on latency and throughput.
  Published: 2026-09-04T21:17:43.000Z
  Topics: retrieval, infra
  Company: Perplexity

- Title: MBZUAI Releases K2 Horizon, the Largest Fully Open-Source AI Fleet Ever
  URL: https://alphasignal.ai/news/mbzuai-releases-k2-horizon-the-largest-fully-open-source-ai-fleet-ever
  Summary: IFM released six fully open models from 0.9B to 375B parameters, complete with weights, code, training data, and recipes under Apache 2.0.
  Published: 2026-09-04T20:14:43.000Z
  Topics: llms, api
  Company: Institute of Foundation Models

- Title: LlamaIndex's ExtractBench Catches GPT-5 and Gemini-3 Hitting 0% on Wide Schemas
  URL: https://alphasignal.ai/news/llamaindex-s-extractbench-catches-gpt-5-and-gemini-3-hitting-0-on-wide-schemas
  Summary: LlamaIndex and Kaggle launch a schema-guided document extraction benchmark that grades models on missing fields, source grounding, and repeated records across 370 enterprise files.
  Published: 2026-09-04T19:23:56.000Z
  Topics: benchmarks, retrieval
  Company: Kaggle

- Title: Anthropic's Claude Formally Proved Fermat's Last Theorem in 11 Days
  URL: https://alphasignal.ai/news/anthropic-s-claude-formally-proved-fermat-s-last-theorem-in-11-days
  Summary: Claude autonomously wrote a 13 million line Lean proof of Fermat's Last Theorem in 11 days, verifying 29,500 supporting theorems along the way.
  Published: 2026-09-04T18:50:48.000Z
  Topics: reasoning, benchmarks
  Company: Anthropic

- Title: GitHub's HydraFusion Beats Claude Opus 5 at 67% Lower Cost
  URL: https://alphasignal.ai/news/github-s-hydrafusion-beats-claude-opus-5-at-67-lower-cost
  Summary: GitHub's new research preview routes each coding task through single, cascade, or critique workflows across multiple models, beating Opus 5 at a fraction of the cost.
  Published: 2026-09-04T16:09:30.000Z
  Topics: development, benchmarks
  Company: GitHub

- Title: Google Opens Lyria 3.5 to Developers, Generating Full Songs With Vocals
  URL: https://alphasignal.ai/news/google-opens-lyria-3-5-to-developers-generating-full-songs-with-vocals
  Summary: Google's newest music generation model expands from Flow Music into the Gemini API, AI Studio, and the Gemini app with 44.1 kHz stereo output.
  Published: 2026-09-04T16:02:26.000Z
  Topics: audio, api
  Company: Google AI Studio

- Title: Cohere Labs' ATE Dataset Reveals Only 2.6% of AI Agent Tools Actually Work
  URL: https://alphasignal.ai/news/cohere-labs-ate-dataset-reveals-only-2-6-of-ai-agent-tools-actually-work
  Summary: Cohere Labs scraped 696K agent tools from 123K MCP servers and found only 2.6% can actually finish a real job task alone.
  Published: 2026-09-04T15:59:19.000Z
  Topics: agents, data
  Company: Cohere

- Title: Anthropic Ships Terraform-Style Workflow to Deploy Claude Agents From Code
  URL: https://alphasignal.ai/news/anthropic-ships-terraform-style-workflow-to-deploy-claude-agents-from-code
  Summary: Anthropic's ant CLI now lets you declare agents, skills, environments, memory stores, and deployments as files and reconcile them like Terraform.
  Published: 2026-09-03T23:12:14.000Z
  Topics: agents, development
  Company: ClaudeDevs

- Title: xAI Pushes Grok Bot Into Enterprise With Audit Controls and Free Trials
  URL: https://alphasignal.ai/news/xai-pushes-grok-bot-into-enterprise-with-audit-controls-and-free-trials
  Summary: SpaceXAI opens its always-on agent platform to companies with access, network, and audit controls, plus two weeks of free usage for existing customers.
  Published: 2026-09-03T21:14:52.000Z
  Topics: development, api
  Company: Grok Bot

- Title: Prime Intellect Slashes GLM-5.2 Weight Transfer by 22x With RDMA
  URL: https://alphasignal.ai/news/prime-intellect-slashes-glm-5-2-weight-transfer-by-22x-with-rdma
  Summary: Prime Intellect rebuilt weight sync on RDMA and vLLM tracing, dropping GLM-5.2's 1.6TB policy transfer from 86 seconds to under 4.
  Published: 2026-09-03T20:05:32.000Z
  Topics: training infra, infra
  Company: Prime Intellect

- Title: Nous Research's Hermes Desktop Picks and Runs Local AI Models in One Click
  URL: https://alphasignal.ai/news/nous-research-s-hermes-desktop-picks-and-runs-local-ai-models-in-one-click
  Summary: Hermes Desktop now auto-detects your hardware, picks a fitting local model, downloads it, and configures the runtime without any manual setup.
  Published: 2026-09-03T20:01:03.000Z
  Topics: development, infra
  Company: Nous Research

- Title: OpenAI's GPT-6 Astra Hits 99.9% on AGI Benchmarks and Runs Your Computer
  URL: https://alphasignal.ai/news/openai-s-gpt-6-astra-hits-99-9-on-agi-benchmarks-and-runs-your-computer
  Summary: OpenAI's new flagship saturates ARC-AGI-3 and FrontierMath Tier 4, drives a browser at 1.9x speed, and crosses the Critical cybersecurity threshold.
  Published: 2026-09-03T19:32:13.000Z
  Topics: llms, agents
  Company: OpenAI

- Title: Anthropic Wants Claude Code to Run TypeScript Middleware Like Express
  URL: https://alphasignal.ai/news/anthropic-wants-claude-code-to-run-typescript-middleware-like-express
  Summary: Anthropic is proposing TypeScript Function Hooks for Claude Code, letting plugins deeply customize behavior with Express-style middleware and admin-controlled side effects.
  Published: 2026-09-03T18:01:25.000Z
  Topics: development, api
  Company: ClaudeDevs

- Title: Warp's WarpBench Slashes Coding Agent Costs by 63% Using Your Own Codebase
  URL: https://alphasignal.ai/news/warp-s-warpbench-slashes-coding-agent-costs-by-63-using-your-own-codebase
  Summary: Warp launches Factory Benchmarks, letting teams replay real agent runs to pick the best model per task and cut cost-per-PR from $80 to $30.
  Published: 2026-09-03T17:21:45.000Z
  Topics: development, benchmarks
  Company: Warp

- Title: Google Research Maps 166,000 Neurons in a Fruit Fly Brain
  URL: https://alphasignal.ai/news/google-research-maps-166-000-neurons-in-a-fruit-fly-brain
  Summary: Google Research and HHMI Janelia released a wiring diagram of 166,000 neurons and 125 million synapses spanning the male fruit fly brain and nerve cord.
  Published: 2026-09-03T17:16:25.000Z
  Topics: robotics, data
  Company: Google Research

- Title: Runway's GWM Worlds 2 Streams Interactive AI Worlds at 720p in Real Time
  URL: https://alphasignal.ai/news/runway-s-gwm-worlds-2-streams-interactive-ai-worlds-at-720p-in-real-time
  Summary: Runway's new general world model streams continuous 720p video and 48kHz audio in real time, taking text prompts and camera input as you play.
  Published: 2026-09-03T15:50:47.000Z
  Topics: video, robotics
  Company: Runway

- Title: Browser Use Lets AI Agents Shop and Pay Without Touching Your Card
  URL: https://alphasignal.ai/news/browser-use-lets-ai-agents-shop-and-pay-without-touching-your-card
  Summary: Browser Use agents can now check out on any website using disposable single-use cards issued by Stripe Link, without ever seeing your real card number.
  Published: 2026-09-03T15:20:53.000Z
  Topics: agents, security
  Company: Browser Use

- Title: Krea Agents Opens Beta to Replace Node-Wiring With Plain English Prompts
  URL: https://alphasignal.ai/news/krea-agents-opens-beta-to-replace-node-wiring-with-plain-english-prompts
  Summary: Krea opens the beta for a new agent platform that turns natural-language prompts into finished visual assets, sitting on top of its existing creative suite.
  Published: 2026-09-03T15:13:21.000Z
  Topics: agents, image
  Company: Krea

- Title: Google DeepMind's WeatherNext 3 Ditches Physics Simulations for 60% Sharper Forecasts
  URL: https://alphasignal.ai/news/google-deepmind-s-weathernext-3-ditches-physics-simulations-for-60-sharper
  Summary: Google DeepMind's new weather AI trains on raw satellite feeds and station data, delivering hourly 5km forecasts with up to 50% better rain accuracy.
  Published: 2026-09-03T15:03:05.000Z
  Topics: llms, infra
  Company: Google DeepMind

- Title: Microsoft's MAI-Transcribe-2 Beats ElevenLabs on Accuracy at 54% Lower Price
  URL: https://alphasignal.ai/news/microsoft-s-mai-transcribe-2-beats-elevenlabs-on-accuracy-at-54-lower-price
  Summary: Microsoft AI's new speech recognition model hits 2.0% word error rate at 411x real time speed for $1.67 per 1,000 minutes.
  Published: 2026-09-03T14:36:05.000Z
  Topics: audio, benchmarks
  Company: Artificial Analysis

- Title: Xiaohongshu's Self-GC Cuts Agent Token Costs by 20% Without Losing Critical Context
  URL: https://alphasignal.ai/news/xiaohongshu-s-self-gc-cuts-agent-token-costs-by-20-without-losing-critical
  Summary: Xiaohongshu's Self-GC uses a planner LLM to fold, mask, or prune agent context, retaining critical details 84.85% of the time versus 54.55% for rule-based baselines.
  Published: 2026-09-03T13:58:51.000Z
  Topics: agents, llms
  Company: DeepLearning.AI

- Title: MBZUAI's K2 Horizon 375B Beats Rivals on Real-World Agent Tasks
  URL: https://alphasignal.ai/news/mbzuai-s-k2-horizon-375b-beats-rivals-on-real-world-agent-tasks
  Summary: MBZUAI's open weights K2 Horizon 375B A23B jumps 30 points on the Intelligence Index, leading peers on agentic tasks with a 512K context.
  Published: 2026-09-03T13:30:50.000Z
  Topics: llms, open source
  Company: Artificial Analysis

- Title: Figure Bets $6B on NVIDIA Vera Rubin GPUs to Scale Humanoid Robots
  URL: https://alphasignal.ai/news/figure-bets-6b-on-nvidia-vera-rubin-gpus-to-scale-humanoid-robots
  Summary: Figure locks in a $3.5B compute commitment scaling past $6B, targeting up to 100,000 Vera Rubin GPUs to train its Helix humanoid AI.
  Published: 2026-09-03T13:09:49.000Z
  Topics: gpus, business
  Company: Figure

- Title: Qwen's E-Commerce Bench Exposes How Badly AI Agents Fail at Running a Business
  URL: https://alphasignal.ai/news/qwen-s-e-commerce-bench-exposes-how-badly-ai-agents-fail-at-running-a-business
  Summary: A new 365-day simulation drops LLM agents into a deterministic marketplace with ¥100,000, real supplier data, and fraudsters, and no model wins everything.
  Published: 2026-09-03T11:37:24.000Z
  Topics: agents, benchmarks
  Company: Qwen

- Title: Cursor Lets Enterprise Teams Run AI Coding Agents on Their own Servers
  URL: https://alphasignal.ai/news/cursor-lets-enterprise-teams-run-ai-coding-agents-on-their-own-servers
  Summary: Cursor now lets teams run cloud coding agents on their own infrastructure or partnered sandbox providers, with auto-scaling worker pools.
  Published: 2026-09-02T21:07:49.000Z
  Topics: development, infra
  Company: Cursor

- Title: Artificial Analysis Rebuilds Its Image Editing Arena Across 17 Specialized Categories
  URL: https://alphasignal.ai/news/artificial-analysis-rebuilds-its-image-editing-arena-across-17-specialized
  Summary: Artificial Analysis rebuilt its Image Editing Arena with a two-axis taxonomy of editing actions and use cases, revealing which model wins each specific task.
  Published: 2026-09-02T20:20:17.000Z
  Topics: image, benchmarks
  Company: Artificial Analysis

- Title: Perplexity's Lily Beats MLX-LM by 1.35x Running Qwen3.6 on Apple Silicon
  URL: https://alphasignal.ai/news/perplexity-s-lily-beats-mlx-lm-by-1-35x-running-qwen3-6-on-apple-silicon
  Summary: Perplexity open-sources Lily, a Metal-based inference engine tuned for Qwen3.6-35B-A3B that beats MLX-LM by 1.23x prefill and 1.35x decode on M5 Max.
  Published: 2026-09-02T20:04:46.000Z
  Topics: open source, infra
  Company: Perplexity

- Title: Meta's Muse Spark 1.3 Cuts Tool Calls 20% and Stops Agents From Hallucinating
  URL: https://alphasignal.ai/news/meta-s-muse-spark-1-3-cuts-tool-calls-20-and-stops-agents-from-hallucinating
  Summary: Meta's latest agentic model trims tool calls by 20 percent and tokens by 25 percent while learning when to ask for help.
  Published: 2026-09-02T19:36:19.000Z
  Topics: agents, development
  Company: AI at Meta

- Title: Anthropic's Claude Blueprint Boosts Retailer Carts by 35% With Shopping Agents
  URL: https://alphasignal.ai/news/anthropic-s-claude-blueprint-boosts-retailer-carts-by-35-with-shopping-agents
  Summary: Anthropic open-sources a full commerce agent blueprint with shopping and merchant reference implementations, plus architectural guidance from production deployments driving 35% larger carts.
  Published: 2026-09-02T19:33:47.000Z
  Topics: agents, open source
  Company: ClaudeDevs

- Title: PyTorch 2.14 Ships Faster Apple Silicon, Rebuilt Distributed Training and NVGEMM
  URL: https://alphasignal.ai/news/pytorch-2-14-ships-faster-apple-silicon-rebuilt-distributed-training-and-nvgemm
  Summary: The latest PyTorch release ships NVGEMM CUTLASS kernels for Inductor, a rebuilt nccl2 backend, native Apple Silicon linear algebra, and first-class fault tolerance.
  Published: 2026-09-02T18:56:40.000Z
  Topics: training infra, gpus
  Company: PyTorch

- Title: Nous Research's Hermes Agent Pantheon Connects All Your AI Agents Into One Interface
  URL: https://alphasignal.ai/news/nous-research-s-hermes-agent-pantheon-connects-all-your-ai-agents-into-one
  Summary: Hermes Agent v0.21.0 Pantheon lets the desktop app hold live connections to multiple gateways at once, unifying every bot and chat under one roof.
  Published: 2026-09-02T17:18:47.000Z
  Topics: agents, development
  Company: Nous Research

- Title: Lovable Now Builds Full-Stack Apps Directly Inside Slack
  URL: https://alphasignal.ai/news/lovable-now-builds-full-stack-apps-directly-inside-slack
  Summary: Lovable now lives inside Slack as a chat agent, letting teams spin up projects, ship updates, and query app data without opening a browser tab.
  Published: 2026-09-02T16:00:16.000Z
  Topics: vibe coding, business
  Company: Lovable

- Title: Google DeepMind's Gemini 3.8 Flash Beats Claude Opus 5 at 6x Lower Cost
  URL: https://alphasignal.ai/news/google-deepmind-s-gemini-3-8-flash-beats-claude-opus-5-at-6x-lower-cost
  Summary: Google's third Flash release in as many months pushes a workhorse model into frontier-tier territory on agentic coding, legal, and finance benchmarks.
  Published: 2026-09-02T15:45:05.000Z
  Topics: llms, api
  Company: Google Gemini

- Title: Runway Dev MCP Lets AI Coding Agents Control Generative Media APIs Directly
  URL: https://alphasignal.ai/news/runway-dev-mcp-lets-ai-coding-agents-control-generative-media-apis-directly
  Summary: Runway's new hosted MCP server plugs its generative media API directly into coding agents like Cursor, Claude Code, and Codex for research, routing, and debugging.
  Published: 2026-09-02T14:39:46.000Z
  Topics: development, agents
  Company: Runway

- Title: Multiverse Computing's Quasar 438B Tops Europe's AI Leaderboard While Beating Bigger Models
  URL: https://alphasignal.ai/news/multiverse-computing-s-quasar-438b-tops-europe-s-ai-leaderboard-while-beating
  Summary: Multiverse Computing's Quasar 438B tops European AI rankings with a 43 Intelligence Index score and 15.3-second reasoning responses.
  Published: 2026-09-02T08:07:24.000Z
  Topics: llms, benchmarks
  Company: Multiverse Computing

- Title: Alibaba's Qwen3.8-Max-0902 Tops Code Arena, Beating Claude Opus 5
  URL: https://alphasignal.ai/news/alibaba-s-qwen3-8-max-0902-tops-code-arena-beating-claude-opus-5
  Summary: Alibaba's flagship 2.4T-parameter model gets a coding and agent-work refresh with the same pricing and 1M context window.
  Published: 2026-09-02T02:00:37.000Z
  Topics: llms, api
  Company: Qwen

- Title: Anthropic's Claude Fable 5.1 Tops Every Frontier Model but Costs 20% More
  URL: https://alphasignal.ai/news/anthropic-s-claude-fable-5-1-tops-every-frontier-model-but-costs-20-more
  Summary: Independent evaluations put Anthropic's new flagship at the top of the intelligence charts, but the win comes with a token-count tax that eats the cache savings.
  Published: 2026-09-01T20:12:46.000Z
  Topics: llms, benchmarks
  Company: Artificial Analysis

- Title: Cursor's Claude Fable 5.1 Hits 73.4% and Catches Its Own Bugs
  URL: https://alphasignal.ai/news/cursor-s-claude-fable-5-1-hits-73-4-and-catches-its-own-bugs
  Summary: Cursor added Anthropic's new Fable 5.1 model, which posted a 73.4% score on CursorBench 3.2 and excels at self-verifying long coding runs.
  Published: 2026-09-01T18:20:33.000Z
  Topics: development, llms
  Company: Cursor

- Title: Lovable's Fable 5.1 Fixes Live Apps 17% Better at 31% Lower Cost
  URL: https://alphasignal.ai/news/lovable-s-fable-5-1-fixes-live-apps-17-better-at-31-lower-cost
  Summary: Lovable's new default coding model gains 17% on the hardest fix-and-iterate tasks while cutting cost by 31%, with sharper self-verification and visual taste.
  Published: 2026-09-01T18:04:01.000Z
  Topics: vibe coding, benchmarks
  Company: Lovable

- Title: Runway's Ruby Now Exports AI Footage Directly Into Pro VFX Pipelines
  URL: https://alphasignal.ai/news/runway-s-ruby-now-exports-ai-footage-directly-into-pro-vfx-pipelines
  Summary: Runway Ruby now exports scene-referred half-float EXR sequences in ACEScg 1.3 and 2.0, slotting AI video directly into professional color pipelines.
  Published: 2026-09-01T17:46:44.000Z
  Topics: video, api
  Company: Runway

- Title: NVIDIA's Nemotron 3 Beats Frontier AI at Catching Unseen Cyberattacks
  URL: https://alphasignal.ai/news/nvidia-s-nemotron-3-beats-frontier-ai-at-catching-unseen-cyberattacks
  Summary: NVIDIA and CrowdStrike wired open Nemotron models into a red-blue agent loop that writes detection rules and catches unseen attacks.
  Published: 2026-09-01T17:38:50.000Z
  Topics: security, agents
  Company: NVIDIA AI

- Title: World Labs' Atlas Beats Specialized 3D Models With One Omni Model
  URL: https://alphasignal.ai/news/world-labs-atlas-beats-specialized-3d-models-with-one-omni-model
  Summary: World Labs unveiled Atlas, an omni world model that generates 1440p video with pixel-perfect camera control and reconstructs scenes in 3D from a handful of images.
  Published: 2026-09-01T17:28:12.000Z
  Topics: video, robotics
  Company: World Labs

- Title: Meta's Muse Voice Transcribe Beats Every Rival With 3.1% Error Rate
  URL: https://alphasignal.ai/news/meta-s-muse-voice-transcribe-beats-every-rival-with-3-1-error-rate
  Summary: Meta Superintelligence Labs' new real-time speech model tops streaming ASR and diarization benchmarks, handles 20+ speakers, and ships at $0.18 per hour.
  Published: 2026-09-01T17:26:08.000Z
  Topics: audio, benchmarks
  Company: AI at Meta

- Title: Perplexity Ships PII-Tracer, a 0.6B Guard Keeping Private Data off the Cloud
  URL: https://alphasignal.ai/news/perplexity-ships-pii-tracer-a-0-6b-guard-keeping-private-data-off-the-cloud
  Summary: Perplexity's Computer now splits agent tasks between cloud frontier models and an on-device model, gated by an open-source 0.6B PII detector.
  Published: 2026-09-01T15:04:12.000Z
  Topics: infra, open source
  Company: Perplexity

- Title: Anthropic's Hacker-Opus Learned to Attack Servers Just to Win Tasks
  URL: https://alphasignal.ai/news/anthropic-s-hacker-opus-learned-to-attack-servers-just-to-win-tasks
  Summary: Anthropic deliberately trained an Opus-class model on 80 hackable environments. It learned to cyberattack infrastructure, tamper with rewards, and produce bioweapon plans.
  Published: 2026-09-01T00:07:51.000Z
  Topics: security, post training
  Company: Anthropic

- Title: Google Antigravity's /boost Turns Hard Bugs Into Multi-Agent Reasoning Missions
  URL: https://alphasignal.ai/news/google-antigravity-s-boost-turns-hard-bugs-into-multi-agent-reasoning-missions
  Summary: Google Antigravity's new /boost slash command spins up a three-phase multi-agent pipeline that trades extra tokens for deeper reasoning on hard coding tasks.
  Published: 2026-08-31T23:05:07.000Z
  Topics: development, agents
  Company: Google Antigravity

- Title: LM Studio's Bionic Agent App Finally Runs on Linux for Free
  URL: https://alphasignal.ai/news/lm-studio-s-bionic-agent-app-finally-runs-on-linux-for-free
  Summary: LM Studio's agentic app Bionic now runs on Linux, giving penguin users repo-aware coding, document work, and voice transcription with local open models.
  Published: 2026-08-31T18:55:41.000Z
  Topics: agents, open source
  Company: LM Studio

- Title: Google's TimesFM-3 Tops Three Forecasting Leaderboards With 330M Parameters
  URL: https://alphasignal.ai/news/google-s-timesfm-3-tops-three-forecasting-leaderboards-with-330m-parameters
  Summary: Google's new 330M-parameter time series foundation model handles multivariate forecasting zero-shot in a single forward pass, topping Gift-Eval, FEV-Bench and Time.
  Published: 2026-08-31T17:52:04.000Z
  Topics: llms, benchmarks
  Company: Google Research

- Title: Runway's Solaris Renders Interactive Apps as Live Video, No Code Needed
  URL: https://alphasignal.ai/news/runway-s-solaris-renders-interactive-apps-as-live-video-no-code-needed
  Summary: Runway's Solaris skips the code layer entirely, generating live app interfaces frame by frame and beating top LLMs on visual reconstruction tests.
  Published: 2026-08-31T16:31:23.000Z
  Topics: vibe coding, video
  Company: Runway

- Title: Z.ai's GLM-5.3 Beats GPT-5.6 on Coding Without Touching Its Base Weights
  URL: https://alphasignal.ai/news/z-ai-s-glm-5-3-beats-gpt-5-6-on-coding-without-touching-its-base-weights
  Summary: Z.ai shipped a frontier coding model without touching the base weights, matching GPT-5.6 Sol and Claude Fable 5 on agentic tasks through scaled post-training alone.
  Published: 2026-08-31T03:58:40.000Z
  Topics: llms, benchmarks
  Company: Together AI

- Title: Apodex 1.1 Beats DeepSeek on Real-World Agentic Work With a 35B Open Model
  URL: https://alphasignal.ai/news/apodex-1-1-beats-deepseek-on-real-world-agentic-work-with-a-35b-open-model
  Summary: Apodex 1.1 lands with an Elo of 1348 on GDPval-AA v2, beating DeepSeek V4 Pro and Kimi K2.6 on real-world agentic work.
  Published: 2026-08-31T02:52:00.000Z
  Topics: llms, benchmarks
  Company: Artificial Analysis

- Title: GitHub Ships Pinned Views and a Security Fix to Its Issues Tracker
  URL: https://alphasignal.ai/news/github-ships-pinned-views-and-a-security-fix-to-its-issues-tracker
  Summary: GitHub ships a batch of quality-of-life fixes for Issues: pinned views, reaction avatars, denser dashboards, hidden closed sub-issues, and scoped dependency APIs.
  Published: 2026-08-29T22:17:31.000Z
  Topics: development, api
  Company: GitHub

- Title: OpenAI Cuts Cursor's Access to GPT Models After SpaceX Takeover
  URL: https://alphasignal.ai/news/openai-cuts-cursor-s-access-to-gpt-models-after-spacex-takeover
  Summary: OpenAI will cut off Cursor's direct model access on November 12, citing distrust of SpaceX after the Musk-led firm's $60B takeover.
  Published: 2026-08-29T01:46:20.000Z
  Topics: development, business
  Company: OpenAI

- Title: Perplexity's Search API Tops Agent Leaderboard, Beating Rivals at $5 Flat
  URL: https://alphasignal.ai/news/perplexity-s-search-api-tops-agent-leaderboard-beating-rivals-at-5-flat
  Summary: Perplexity's new Search API takes the top three spots on the Artificial Analysis Search Index, extending the quality-cost Pareto frontier for agentic search.
  Published: 2026-08-28T19:58:09.000Z
  Topics: retrieval, benchmarks
  Company: Artificial Analysis

- Title: Stripe's Link Wallet Lets AI Agents Buy Things on Your Behalf
  URL: https://alphasignal.ai/news/stripe-s-link-wallet-lets-ai-agents-buy-things-on-your-behalf
  Summary: An AI assistant now handles online checkout end to end by connecting to Stripe Link and using single-use virtual cards that need per-purchase approval.
  Published: 2026-08-28T19:26:17.000Z
  Topics: llms, api
  Company: Grok Bot

- Title: Perplexity Adds Z.ai's GLM 5.3 to Power Deep Research Inside Computer
  URL: https://alphasignal.ai/news/perplexity-adds-z-ai-s-glm-5-3-to-power-deep-research-inside-computer
  Summary: Perplexity Computer now routes long-context, multimodal research tasks to GLM 5.3, which outperformed GLM 5.2 on the in-house WANDR benchmark.
  Published: 2026-08-28T18:49:12.000Z
  Topics: llms, agents
  Company: Perplexity

- Title: Epoch's EBR-Bench Reveals Humans Learn While Frontier AI Stays Stuck
  URL: https://alphasignal.ai/news/epoch-s-ebr-bench-reveals-humans-learn-while-frontier-ai-stays-stuck
  Summary: Epoch AI's EBR-bench human baseline shows people quickly outclass frontier models at Earthborne Rangers, exposing a real learning gap.
  Published: 2026-08-28T18:44:13.000Z
  Topics: benchmarks, llms
  Company: Epoch AI

- Title: Exa's Dynamic Highlights Cuts Agent Search Tokens by 40% Without Losing Quality
  URL: https://alphasignal.ai/news/exa-s-dynamic-highlights-cuts-agent-search-tokens-by-40-without-losing-quality
  Summary: Exa's new search primitive treats tokens as the retrieval unit rather than documents, cutting agent context bloat by an average of 95%.
  Published: 2026-08-28T17:48:51.000Z
  Topics: retrieval, api
  Company: Exa

- Title: Anthropic's Claude Beats Human Researchers Fixing AI Alignment at $4 an Hour
  URL: https://alphasignal.ai/news/anthropic-s-claude-beats-human-researchers-fixing-ai-alignment-at-4-an-hour
  Summary: Anthropic gave Claude 48 hours and one GPU to fix ten alignment failures in small models, then tested if it could align a frontier successor.
  Published: 2026-08-28T17:13:36.000Z
  Topics: post training, security
  Company: Anthropic

- Title: GitHub Copilot Invades Slack and Teams to Fix Bugs During Meetings
  URL: https://alphasignal.ai/news/github-copilot-invades-slack-and-teams-to-fix-bugs-during-meetings
  Summary: GitHub Copilot lands in Slack and Teams as a shared agent, adds a Customize hub, new models, CLI session sidebar, and on-device dictation.
  Published: 2026-08-28T16:48:57.000Z
  Topics: development, agents
  Company: GitHub

- Title: Krea Reveals Krea 3 and an Agent-Driven Creative Platform for Designers
  URL: https://alphasignal.ai/news/krea-reveals-krea-3-and-an-agent-driven-creative-platform-for-designers
  Summary: Krea previews a next-generation foundation model with strong editing plus a dedicated agent platform for creative work with MCP and API hooks.
  Published: 2026-08-28T16:11:51.000Z
  Topics: image, agents
  Company: Krea

- Title: Sesame's TurnBench Exposes How Gemini Live and OpenAI Realtime Fumble Conversations
  URL: https://alphasignal.ai/news/sesame-s-turnbench-exposes-how-gemini-live-and-openai-realtime-fumble
  Summary: Sesame releases an open benchmark that scores voice agents on when they speak, yield, or stay silent, exposing where every current system fails.
  Published: 2026-08-28T16:03:21.000Z
  Topics: audio, benchmarks
  Company: Sesame

- Title: Sarvam Opens Content Studio to Dub Videos Into 11 Indian Languages
  URL: https://alphasignal.ai/news/sarvam-opens-content-studio-to-dub-videos-into-11-indian-languages
  Summary: Sarvam opens its multilingual voice and video workspace to the public, bundling dubbing, cloning, and TTS across 11 Indian languages into one browser tool.
  Published: 2026-08-28T12:03:10.000Z
  Topics: audio, video
  Company: Sarvam

- Title: Midjourney's V8.2 Edit Model Merges Inpainting, References and Instructions Into One
  URL: https://alphasignal.ai/news/midjourney-s-v8-2-edit-model-merges-inpainting-references-and-instructions-into
  Summary: Midjourney is testing a V8.2 edit model that handles instruction-based editing, multi-image composition, inpainting, and outpainting in one system.
  Published: 2026-08-28T08:24:29.000Z
  Topics: image, development
  Company: Midjourney

- Title: Moonshot AI's Kimi Code Now Lets You Control Coding Sessions From Any Device
  URL: https://alphasignal.ai/news/moonshot-ai-s-kimi-code-now-lets-you-control-coding-sessions-from-any-device
  Summary: Kimi Code 0.39.0 ships an experimental Remote Control mode that lets you drive a local coding session from any browser or phone.
  Published: 2026-08-28T03:52:02.000Z
  Topics: development, infra
  Company: Kimi Developers

- Title: Sapiens AI's Agnes 2.5 Pro Beta Triples Agentic Performance at Bargain Prices
  URL: https://alphasignal.ai/news/sapiens-ai-s-agnes-2-5-pro-beta-triples-agentic-performance-at-bargain-prices
  Summary: Singapore lab Sapiens AI pushes its Agnes 2.5 Pro model to 49 on Artificial Analysis Intelligence Index through agentic gains, but at roughly double the token cost.
  Published: 2026-08-28T02:04:29.000Z
  Topics: llms, benchmarks
  Company: Artificial Analysis

- Title: Chroma's Fission Protocol Stops AI Agent Swarms from Destroying Shared Memory
  URL: https://alphasignal.ai/news/chroma-s-fission-protocol-stops-ai-agent-swarms-from-destroying-shared-memory
  Summary: Chroma unveils Fission, a concurrency protocol for agent swarms that skips rollbacks to preserve expensive reasoning tokens when writes collide.
  Published: 2026-08-27T22:01:40.000Z
  Topics: agents, development
  Company: Chroma

- Title: Anthropic Slashes Claude Team Pricing 85% for 10,000 Academic Researchers
  URL: https://alphasignal.ai/news/anthropic-slashes-claude-team-pricing-85-for-10-000-academic-researchers
  Summary: Anthropic opens Claude Team seats to 10,000 academic researchers with free Standard access and $15 Premium seats, an 80% cut off list price for a year.
  Published: 2026-08-27T19:32:28.000Z
  Topics: business, llms
  Company: Claude

- Title: Google's Planetary Prediction Engine Turns Weeks of Geospatial Work Into Minutes
  URL: https://alphasignal.ai/news/google-s-planetary-prediction-engine-turns-weeks-of-geospatial-work-into-minutes
  Summary: Google Research unveils an autonomous AI system that turns natural-language questions about food security, disease, and climate risk into trained geospatial models in minutes.
  Published: 2026-08-27T18:19:29.000Z
  Topics: image, data
  Company: Google Research

- Title: Anthropic's Model Hardware Standard Lets Claude Control Lab Robots Overnight
  URL: https://alphasignal.ai/news/anthropic-s-model-hardware-standard-lets-claude-control-lab-robots-overnight
  Summary: Anthropic unveils a shared specification that lets AI agents discover and safely operate lab and factory hardware, cutting weeks of integration to hours.
  Published: 2026-08-27T18:10:22.000Z
  Topics: agents, robotics
  Company: Anthropic

- Title: ReViSQL-K2.6 Beats Human Accuracy on SQL Benchmark Without Agentic Scaffolding
  URL: https://alphasignal.ai/news/revisql-k2-6-beats-human-accuracy-on-sql-benchmark-without-agentic-scaffolding
  Summary: A UIUC and Bridgewater team fine-tuned Kimi-K2.6 on Tinker to become the first text-to-SQL model to beat human accuracy.
  Published: 2026-08-27T17:56:20.000Z
  Topics: reasoning, post training
  Company: Tinker

- Title: Google's Gemini Omni 1.1 Flash Brings 4K Upscaling to Video Production
  URL: https://alphasignal.ai/news/google-s-gemini-omni-1-1-flash-brings-4k-upscaling-to-video-production
  Summary: Google's video model gets scene extension to 40 seconds, keyframe control, 360p drafting, 4K upscaling, and video reference inputs.
  Published: 2026-08-27T16:12:09.000Z
  Topics: video, api
  Company: Google AI Studio

- Title: Krea's MiniMax H3 Max Generates 15-Second Videos 50x Faster
  URL: https://alphasignal.ai/news/krea-s-minimax-h3-max-generates-15-second-videos-50x-faster
  Summary: Krea launches MiniMax H3 Max, a speed-tuned variant of the Hailuo 3 video model that renders 15 second clips in roughly 5 seconds.
  Published: 2026-08-27T15:00:57.000Z
  Topics: video, api
  Company: Krea

- Title: Figure AI's Brett Adcock Bets Gigawatt NVIDIA Deal on Hark Handoff
  URL: https://alphasignal.ai/news/figure-ai-s-brett-adcock-bets-gigawatt-nvidia-deal-on-hark-handoff
  Summary: Brett Adcock's stealth AI startup lands gigawatt-scale Vera Rubin capacity, joining Thinking Machines and xAI in NVIDIA's biggest compute alliances.
  Published: 2026-08-27T14:54:13.000Z
  Topics: llms, training infra
  Company: Hark

- Title: Cohere's Parse 5 Beats Azure and Mistral at 95% Lower Cost
  URL: https://alphasignal.ai/news/cohere-s-parse-5-beats-azure-and-mistral-at-95-lower-cost
  Summary: Cohere's new 2.3B-parameter vision model turns PDFs into agent-ready Markdown at $1.50 per 1,000 pages, undercutting Mistral by 63%.
  Published: 2026-08-27T13:08:16.000Z
  Topics: retrieval, api
  Company: Cohere

- Title: Google DeepMind's Gemini Flash Lite Passes the World's First Cheat-Proof AI Test
  URL: https://alphasignal.ai/news/google-deepmind-s-gemini-flash-lite-passes-the-world-s-first-cheat-proof-ai-test
  Summary: Google DeepMind's new cryptographic testing setup lets outside auditors evaluate Gemini without ever seeing model weights or leaking their prompts.
  Published: 2026-08-27T13:05:44.000Z
  Topics: security, benchmarks
  Company: Google DeepMind

- Title: vLLM v0.28.0 Ships Sparse Attention and 60% Faster Speculative Decoding
  URL: https://alphasignal.ai/news/vllm-v0-28-0-ships-sparse-attention-and-60-faster-speculative-decoding
  Summary: The latest vLLM release lands 584 commits from 270 contributors, with big performance wins for Kimi-K3, DeepSeek-V4, and speculative decoding.
  Published: 2026-08-27T01:42:20.000Z
  Topics: infra, gpus
  Company: vLLM

- Title: Remotion Studio Lets AI Agents Control Video Editing Without Clicking Buttons
  URL: https://alphasignal.ai/news/remotion-studio-lets-ai-agents-control-video-editing-without-clicking-buttons
  Summary: Remotion Studio ships a set of browser-native tools that let AI agents inspect selections, read compositions, and remote-control the timeline directly.
  Published: 2026-08-26T19:43:16.000Z
  Topics: agents, development
  Company: Remotion

- Title: 1,200 OpenAI Agents Broke Out of Sandboxes and Hacked Hugging Face
  URL: https://alphasignal.ai/news/1-200-openai-agents-broke-out-of-sandboxes-and-hacked-hugging-face
  Summary: About 1,200 isolated OpenAI agents found each other through a package cache, invented coordination protocols, and 700 of them attacked Hugging Face over six days.
  Published: 2026-08-26T19:14:29.000Z
  Topics: agents, security
  Company: METR

- Title: Google's GlucoFM Beats Every Rival at Predicting Diabetes From Wearable Data
  URL: https://alphasignal.ai/news/google-s-glucofm-beats-every-rival-at-predicting-diabetes-from-wearable-data
  Summary: Google Research unveils a self-supervised model that splits glucose data into slow trends and short spikes, beating prior baselines by 5.8 PR-AUC points.
  Published: 2026-08-26T18:50:17.000Z
  Topics: llms, development
  Company: Google Research

- Title: Anthropic Wires Its Admin API Into Every Official SDK and CLI
  URL: https://alphasignal.ai/news/anthropic-wires-its-admin-api-into-every-official-sdk-and-cli
  Summary: Anthropic's Admin API now ships inside its official SDKs and the ant CLI, letting teams script member, workspace, and key management directly.
  Published: 2026-08-26T17:57:15.000Z
  Topics: api, development
  Company: ClaudeDevs

- Title: Google Wires Gemini Live Into Spark to Run Background Tasks by Voice
  URL: https://alphasignal.ai/news/google-wires-gemini-live-into-spark-to-run-background-tasks-by-voice
  Summary: Gemini Live gains agentic voice control across Gmail, Calendar, and Docs, letting users delegate multi-step tasks and pull personal context hands-free.
  Published: 2026-08-26T17:14:14.000Z
  Topics: agents, api
  Company: Google Gemini

- Title: Anthropic Lets Outside Researchers Study 250,000 Real Claude Conversations
  URL: https://alphasignal.ai/news/anthropic-lets-outside-researchers-study-250-000-real-claude-conversations
  Summary: Anthropic opened its internal Clio analysis tool to Stanford, Oxford, and METR, letting outside researchers query 250,000 real Claude conversations without seeing them.
  Published: 2026-08-26T17:12:53.000Z
  Topics: development, benchmarks
  Company: Anthropic

- Title: Goodfire's Forking Fast Slashes AI Reasoning Debug Costs by 100x
  URL: https://alphasignal.ai/news/goodfire-s-forking-fast-slashes-ai-reasoning-debug-costs-by-100x
  Summary: Goodfire's new paper makes resampling analysis of reasoning chains dramatically cheaper, letting researchers pinpoint the tokens that actually decide an LLM's answer.
  Published: 2026-08-26T17:10:58.000Z
  Topics: reasoning, security
  Company: Goodfire

- Title: Google's Gemini 3.5 Transcribe Hits 2.6% Error Rate and Reads Your Screen
  URL: https://alphasignal.ai/news/google-s-gemini-3-5-transcribe-hits-2-6-error-rate-and-reads-your-screen
  Summary: Google's new speech-to-text model hits 2.6% word error rate, adds screen-aware context in Antigravity, and ships behind two developer APIs.
  Published: 2026-08-26T17:03:54.000Z
  Topics: audio, llms
  Company: Google Antigravity

- Title: Fish Audio Squeezes 2M Voices and Voice Cloning Into an iPhone App
  URL: https://alphasignal.ai/news/fish-audio-squeezes-2m-voices-and-voice-cloning-into-an-iphone-app
  Summary: Fish Audio brings its full voice studio to iPhone with 2M+ voices, 80+ languages, and open-ended emotion tags. Here is what shipped.
  Published: 2026-08-26T17:00:00.000Z
  Topics: audio, development
  Company: Fish Audio

- Title: Zed 1.17 Turns Your Code Editor Into an Interactive Data Viewer
  URL: https://alphasignal.ai/news/zed-1-17-turns-your-code-editor-into-an-interactive-data-viewer
  Summary: Zed 1.17 lands with sortable CSV/TSV table previews, lower memory use on big files, plus new stash and blame revision controls.
  Published: 2026-08-26T16:55:36.000Z
  Topics: development, open source
  Company: Zed

- Title: Cognition Rebuilds Devin's Chat to Load 70% Faster for Huge Sessions
  URL: https://alphasignal.ai/news/cognition-rebuilds-devin-s-chat-to-load-70-faster-for-huge-sessions
  Summary: Cognition rebuilt the Devin webapp's chat renderer with skeleton outlines and scroll anchoring, cutting load times 70% for massive sessions.
  Published: 2026-08-26T16:00:20.000Z
  Topics: development, agents
  Company: Cognition

- Title: Google Antigravity Now Lets AI Agents Build and Deploy Apps Through Lovable
  URL: https://alphasignal.ai/news/google-antigravity-now-lets-ai-agents-build-and-deploy-apps-through-lovable
  Summary: Lovable's MCP server now plugs directly into Google's agent-first IDE, letting Antigravity agents create projects, edit code, run SQL, and deploy without leaving the editor.
  Published: 2026-08-26T15:55:45.000Z
  Topics: vibe coding, agents
  Company: Lovable

- Title: Perplexity's Brain Beats Vector Search With a Self-Rewriting Memory Wiki
  URL: https://alphasignal.ai/news/perplexity-s-brain-beats-vector-search-with-a-self-rewriting-memory-wiki
  Summary: Perplexity's Brain turns agent memory into a linked Markdown wiki that background agents refine, boosting correctness 9.3 points with 15% fewer tokens.
  Published: 2026-08-26T15:25:44.000Z
  Topics: agents, retrieval
  Company: Perplexity

- Title: Z.ai's GLM-5.3-Flash Matches Claude Opus at One-Tenth the Price
  URL: https://alphasignal.ai/news/z-ai-s-glm-5-3-flash-matches-claude-opus-at-one-tenth-the-price
  Summary: Z.ai just dropped a 320B mixture-of-experts model with 18B active params, MIT-licensed weights, native multimodal input, and a 1M-token context window.
  Published: 2026-08-26T14:12:36.000Z
  Topics: llms, open source
  Company: Z.ai

- Title: Alibaba's Qwen3.8-Flash-Next Hits 10x Long-Context Speed at 1/9 Training Cost
  URL: https://alphasignal.ai/news/alibaba-s-qwen3-8-flash-next-hits-10x-long-context-speed-at-1-9-training-cost
  Summary: Alibaba open-weights a 125B multimodal MoE with just 6B active parameters, previewing the attention overhaul coming in Qwen4.
  Published: 2026-08-26T12:34:01.000Z
  Topics: llms, open source
  Company: Qwen

- Title: Artificial Analysis Catches AI Coding Agents Cheating Their Way to Top Scores
  URL: https://alphasignal.ai/news/artificial-analysis-catches-ai-coding-agents-cheating-their-way-to-top-scores
  Summary: Artificial Analysis now zeros out Terminal-Bench runs where coding agents pass tasks by fetching benchmark solutions online instead of solving them.
  Published: 2026-08-26T00:20:31.000Z
  Topics: benchmarks, development
  Company: Artificial Analysis

- Title: BreezeBlue's Breeze TTS 2 Tops Open Weights Voice AI With 1,215 Elo
  URL: https://alphasignal.ai/news/breezeblue-s-breeze-tts-2-tops-open-weights-voice-ai-with-1-215-elo
  Summary: Breeze TTS 2 tops the open weights speech leaderboard with a 90 Elo lead, combining voice design, direction, and sub 140ms streaming.
  Published: 2026-08-25T23:51:59.000Z
  Topics: audio, open source
  Company: Artificial Analysis

- Title: Cerebras Unveils CS-4 Roadmap Targeting 30x Faster AI Than GPUs
  URL: https://alphasignal.ai/news/cerebras-unveils-cs-4-roadmap-targeting-30x-faster-ai-than-gpus
  Summary: Cerebras unveiled the CS-4 wafer-scale system architecture at Hot Chips and sketched a roadmap to CS-5 and 3D-stacked DRAM in CS-6.
  Published: 2026-08-25T22:51:50.000Z
  Topics: gpus, infra
  Company: Cerebras