News · Leaderboard

Latest news, beside the leaderboard

Read what just changed, and see who is ahead. Those are the two doors on this page.

O
OpenAI@OpenAI·Collected 1h agoFresh · 6h
📊 Benchmark

OpenAI Shows Its Models Designing Chips: Up to 14% PPA Gain on AMD Flagship GPU IP, 10–13% Area Savings on Jalapeño, and GPT-Synopsys Built with Synopsys

On November 18 OpenAI published results from using its frontier models for chip design. Working with AMD, its model rewrote SystemVerilog RTL and found up to 14% PPA (power, performance, area) improvement on AMD's heavily optimized flagship GPU IP in under five months, and AMD has already put some changes into a design sent for manufacturing. On OpenAI's own Jalapeño inference chip, model-proposed changes cut standard-cell area by about 13% in the warp engine and 10% in the matrix unit, proven in silicon. OpenAI also shared RISC-V core evals where GPT-6.2 Astra scored highest, and said it is training a new model, GPT-Synopsys, with Synopsys.

OpenAI Shows Its Models Designing Chips: Up to 14% PPA Gain on AMD Flagship GPU IP, 10–13% Area Savings on Jalapeño, and GPT-Synopsys Built with Synopsys
⚡ Key Takeaways
  • •AMD: up to 14% PPA improvement on the flagship GPU IP versus heavily optimized baselines, which OpenAI calls roughly a year of progress in under five months; some changes are already in a design AMD sent for manufacturing
  • •AMD also verified single-digit percentage PPA gains across key blocks of its upcoming next-generation flagship CPU
  • •Jalapeño: model-proposed changes cut standard-cell area by about 13% in the warp engine and 10% in the matrix unit; one multiplier used about 15% less area while dropping latency from two cycles to one, all silicon proven
  • •RISC-V evals: on the open-source PicoRV32 and Ibex cores, GPT-6.2 Astra had the highest average score at about 112,000 tokens per run, across 9,881 scored runs
  • •Synopsys: the companies are training GPT-Synopsys to use Synopsys EDA tools like expert chip designers; rewrites were checked for functional equivalence with Synopsys tools
Read details→
M
Mistral AI@MistralAI·Collected 2h agoFresh · 6h
🛠️ Tooling

Mistral Launches Vibe Desktop for macOS and Windows with Parallel Coding Agents, Diffs and PR Tracking in One Window; Developers in Europe Get 3 Free Months of Vibe Pro with Code LECHONK

On November 17 Mistral announced Vibe desktop on X: hand Vibe several tasks at once, watch the agents work side by side, then review their diffs and track pull requests in the same window, with agents running locally or remotely on macOS and Windows. Developers in Europe can get three free months of Vibe Pro with the code LECHONK, first come, first served; the codes ran out on day one and Mistral added more on November 18. The official site lists Pro at $14.99 per month excluding taxes.

Mistral Launches Vibe Desktop for macOS and Windows with Parallel Coding Agents, Diffs and PR Tracking in One Window; Developers in Europe Get 3 Free Months of Vibe Pro with Code LECHONK
⚡ Key Takeaways
  • •Desktop: available on macOS and Windows; hand Vibe several tasks at once, watch agents work side by side, and review diffs and track pull requests in one window
  • •How it runs: agents can run locally or remotely; the official site says Vibe also ships a CLI and works in VS Code, JetBrains and Zed through ACP
  • •Offer: developers in Europe get three free months of Vibe Pro with code LECHONK, first come, first served, terms apply; codes ran out on day one and more were added on November 18
  • •Pricing: the site lists a Free tier, Pro at $14.99 per month and a team tier at $24.99 per month, all excluding taxes
  • •Model: Mistral says Vibe with Le Chonk scores highest among open models on FrontierCode and Terminal-Bench; the posts do not give the exact scores
Read details→
N
NVIDIA Newsroom@nvidianewsroom·Collected 3h agoFresh · 6h
🚀 Release

NVIDIA Q3 Fiscal 2027: Revenue of $115.8 Billion, Up 103%, Data Center at $107.6 Billion, Vera Rubin NVL72 Now Shipping, Q4 Guidance of $130 Billion

On November 17 NVIDIA reported results for the third quarter of fiscal 2027, ended October 25: revenue of $115.8 billion, up 20% from the prior quarter and 103% from a year ago; Data Center revenue of $107.6 billion, up 110%; GAAP gross margin of 74.8%. Vera Rubin NVL72 has shipped and is running at CoreWeave, Google Cloud, Azure, OCI and Nebius. Q4 revenue guidance is $130.0 billion, plus or minus 3%, with no China data center compute revenue assumed. It is the first report issued under interim CEO Colette Kress.

⚡ Key Takeaways
  • •Revenue: Q3 revenue was $115.8 billion, up 20% quarter over quarter and 103% year over year; GAAP net income was $65.3 billion, or $2.70 per diluted share
  • •Data Center: revenue of $107.6 billion, up 21% sequentially and 110% from a year ago, more than 90% of the total; Edge Computing revenue was $8.2 billion
  • •Vera Rubin: NVL72 has shipped in what NVIDIA calls the fastest product ramp in its history, with racks running at CoreWeave, Google Cloud, Microsoft Azure, OCI and Nebius
  • •Outlook: Q4, which includes an extra week, is guided at $130.0 billion plus or minus 3% with 71% to 72% gross margin, assuming no China data center compute revenue
  • •Also: NVIDIA returned $28.1 billion to shareholders in buybacks and dividends, and the release lists the agreement to acquire Hugging Face and the Open Agent Safety Platform among quarter highlights
Read details→
C
Cognition@cognition·Collected 9h ago
🛠️ Tooling

Cognition Launches Devin Bots: A Persistent Bot Leads a Team of Devins in Parallel, Talks Over Slack, Text and Phone, and Beta Testers Merge 4.5x More PRs per Message

Cognition announced Devin Bots on X on November 17. You give a Bot a goal; it breaks the work down, starts a Devin session for each piece, and runs them in parallel, each on its own cloud machine, through coding, testing and review until the PR is ready to merge, surfacing only the decisions that need you. Bots can run on schedules and triggers, such as watching Slack for customer issues or checking CI every night. Cognition says beta testers merge 4.5x more PRs per message on average.

Cognition Launches Devin Bots: A Persistent Bot Leads a Team of Devins in Parallel, Talks Over Slack, Text and Phone, and Beta Testers Merge 4.5x More PRs per Message
⚡ Key Takeaways
  • •How it works: give a Devin Bot a goal and it breaks the work down and runs several Devin sessions in parallel, each on its own cloud machine, following each one until the PR is ready to merge
  • •Communication: talk to Bots over Slack, group chats, text or phone calls; deploy approvals, missing keys and trade-off decisions land in a single "Needs you" list
  • •Always on: Bots can set schedules and triggers, such as watching Slack for customer issues, checking CI nightly or monitoring Linear for blockers; Bots also hand work to each other, for example a QA Bot passing a bug to a Developer Bot
  • •Numbers: Cognition says beta testers merge 4.5x more PRs per message on average and that some engineers went from 50 PRs a week to nearly 2,000; these are Cognition's own figures
  • •Templates: the site offers Software Factory, Android Engineer, Chief of Staff, Latency Optimizer, Product Engineer and QA Engineer templates; separate pricing has not been published
Read details→
ADSponsored
T
Tibo (Codex & ChatGPT, OpenAI)@thsottiaux·Collected 10h ago
🚀 Release

X Discussion: OpenAI's Tibo Responds to Split Reactions on GPT-6.2, Saying Early Reviewers Tested an Older Build and the Models Changed a Lot in the Final Days

After GPT-6.2 Astra and Sol launched, reactions from influencers and from everyday users diverged sharply. Tibo (@thsottiaux), who works on Codex and ChatGPT at OpenAI, explained on X that a few people got early access a while ago and recorded their videos last week, while the models changed a lot in the last 3 to 4 days, and he believes those testers did not re-test. He also said the models are by no means perfect and that taste and ease of use will improve in future iterations. The post had about 7,300 likes and more than 1,200 replies; what follows is his personal account.

⚡ Key Takeaways
  • •Background: GPT-6.2 Astra and Sol are live in ChatGPT, Codex and the API, and since launch reactions from influencers and everyday users have clearly diverged
  • •Explanation: Tibo says a few people got early access a while ago and recorded their videos last week, but the models changed a lot in the last 3 to 4 days, and he believes nobody re-tested in that window
  • •Stance: he calls the models a hefty jump over their GPT predecessors but by no means perfect, and says the teams want future iterations to improve taste and ease of use
  • •Reach: posted at 13:47 SGT on November 18, with about 7,300 likes, 1,278 replies and roughly 500,000 views
  • •This is a personal account by an OpenAI employee, not an official announcement; OpenAI has not said what exactly changed
Read details→
M
Morteza Mardani (Principal Scientist, NVIDIA)@MardaniMorteza·Collected 14h ago
🚀 Release

NVIDIA Open-Sources Nemotron-Labs-Sigma: Continuous Diffusion Language Models at 3B and 8B, 14 Checkpoints, Block-Parallel Decoding, Commercial Use Allowed

NVIDIA Research has released the Nemotron-Labs-Sigma family on Hugging Face: 14 checkpoints at 3B and 8B, covering base, instruct, few-step distilled and self-speculative decoding variants. Instead of generating one token at a time, the models denoise Gaussian noise in a 16-dimensional continuous embedding space and emit text in parallel blocks. NVIDIA Principal Scientist Morteza Mardani announced the release on X on November 16. The weights use the NVIDIA Nemotron Open Model License and allow commercial use.

NVIDIA Open-Sources Nemotron-Labs-Sigma: Continuous Diffusion Language Models at 3B and 8B, 14 Checkpoints, Block-Parallel Decoding, Commercial Use Allowed
⚡ Key Takeaways
  • •Scale: 14 checkpoints at 3B and 8B, including Base, Instruct, distill-nfe2/4/8/16 and distill-spec variants, all under the nvidia account on Hugging Face
  • •Method: diffusion denoising in a 16-dimensional continuous token-embedding space, generating blocks of tokens (32 to 64 by default) in parallel while reusing committed tokens through a standard KV cache
  • •Speed: NVIDIA research director Arash Vahdat says on X that the distilled models stay strong with just 4 sampling steps and that speculative decoding is 3.5x faster
  • •The distill-spec models draft a 32-token block with the diffusion head and verify it with their own autoregressive head, so output matches greedy autoregressive decoding
  • •Limits: up to 8,192 tokens of context (4,096 for the spec models), English only, NVIDIA GPU in bfloat16 required, no CPU inference
Read details→
K
Kling AI@Kling_ai·Collected 16h ago
🚀 Release

Kling AI Launches Kling Image 4.0: Native 8K Generation, Sharper Text Rendering, Coherent Image Sets From One Prompt, Long-Image Generation and Layered Editing, Now Available

Kling AI, the generative AI brand of Kuaishou, officially introduced the image model Kling Image 4.0 on X on November 17 and announced on November 18 that it is now available. The official site lists five headline capabilities: true-to-life realism with natural lighting, native 8K resolution, crisp small text and creative typography, coherent image sets generated from a single prompt, and long-image generation with layered editing. Kling has not published benchmarks, pricing or API plans.

⚡ Key Takeaways
  • •Timeline: teased as "coming soon" on November 16, introduced on November 17, and announced as available on November 18
  • •Image quality: the official site stresses natural lighting and lifelike textures, with native 8K generation that keeps detail when zoomed in
  • •Text: Kling says text rendering is stronger, handling crisp small text and creative typography for posters, packaging and ads
  • •Batch output: one prompt can produce a coherent series of images; long-image generation and editing (full-page view, layered editing) is also supported
  • •No benchmarks, pricing or API details have been published; it can be tried on the klingai.com web app
Read details→
J
Jensen Huang@JensenHuang·Collected 19h ago
🚀 Release

X Discussion: Jensen Huang Says He Would Be Honored to Return as NVIDIA CEO if the Board Decides It Is in the Best Interest of Shareholders, Employees, the Company and America

At 00:18 Singapore time on November 18 (16:18 UTC on November 17), NVIDIA founder Jensen Huang posted a long statement on his verified account @JensenHuang, his first public remarks since leaving the company. He wrote that he "did not live up to the standards expected of everyone at NVIDIA," has "made serious mistakes in my personal life," and "created problems for the Board" that he deeply regrets, and that he is committed to cooperating with the Board's independent investigation. He also said that since his departure he has seen calls from employees, investors, business partners and customers for him to return as CEO, and that if the Board of Directors decides his return is in the best interest of shareholders, employees, the company and America, he "would be honored to answer these calls and return as CEO." At the time of writing the post had about 830,000 likes and 31 million views. This is his personal statement, not a company decision. The official background is NVIDIA's November 12 announcement that Huang resigned as president, CEO and a member of the Board, effective immediately, with CFO Colette Kress appointed interim CEO.

⚡ Key Takeaways
  • •Jensen Huang posted a long statement on his verified account @JensenHuang at 00:18 Singapore time on November 18; at the time of writing it had about 830,000 likes and 31 million views (post)
  • •He wrote that if the Board decides his return is in the best interest of shareholders, employees, the company and America, he "would be honored to answer these calls and return as CEO" — that is a personal offer, not a company appointment
  • •He also wrote that he has "made serious mistakes in my personal life" and "created problems for the Board," and that he is committed to cooperating with the Board's independent investigation
  • •Official background: on November 12 NVIDIA announced that he resigned as president, CEO and a Board member, effective immediately, with CFO Colette Kress appointed interim CEO (newsroom)
  • •On November 17 the newsroom announced a new National Security and Export Control Oversight Committee, which is separate from whether he returns (newsroom)
Read details→
ADSponsored
M
Microsoft AI@MicrosoftAI·Collected 19h ago
🚀 Release

Microsoft Launches Its First Speech-to-Speech Model, MAI Realtime 1: From $0.01 per Audio Minute, Beats GPT Realtime 2.1 on Quality, Speed and Price per Microsoft, OpenAI Realtime API Compatible, Limited Preview in Foundry

Microsoft AI (MAI) released MAI Realtime 1, its first speech-to-speech model, during Ignite on November 17. It targets enterprise voice agents, with base speech-to-speech usage at $0.01 per minute and reasoning billed separately at $1 per million input tokens and $5 per million output tokens. Microsoft says it beats GPT Realtime 2.1 on quality, speed and price across public benchmarks such as Big Bench Audio and tau-Voice and on Microsoft product evaluations. The model is compatible with the OpenAI Realtime API for easy migration; it can be tried in MAI Labs, and limited-preview customers in Foundry are not billed until public preview.

Microsoft Launches Its First Speech-to-Speech Model, MAI Realtime 1: From $0.01 per Audio Minute, Beats GPT Realtime 2.1 on Quality, Speed and Price per Microsoft, OpenAI Realtime API Compatible, Limited Preview in Foundry
⚡ Key Takeaways
  • •Pricing: base speech-to-speech usage is $0.01 per minute; reasoning is billed separately at $1 per million input tokens and $5 per million output tokens
  • •Benchmarks: Microsoft says it sits on the performance-price Pareto frontier for Big Bench Audio speech reasoning and tau-Voice agentic task completion, and beats GPT Realtime 2.1 on quality, speed and price; the blog gives no exact scores
  • •Capabilities: handles interruptions and quick acknowledgments and knows when to wait; a fast system keeps the conversation flowing while a stronger reasoning engine handles harder tasks; it copes with background noise and connects to enterprise systems and tools
  • •Compatible with the OpenAI Realtime API, so existing voice apps can be tested and migrated with little work
  • •Availability: try it in MAI Labs or join the limited-access preview in Microsoft Foundry (no billing until public preview); Agoda and Reservations.ai are testing it, and it is coming to Dynamics 365 Contact Center and Dragon Copilot
Read details→
L
Logan Kilpatrick@OfficialLoganK·Collected 19h ago
🛠️ Tooling

Google AI Studio Builds Full-Stack Android Apps From a Prompt: The Build Agent Generates Kotlin Apps With Firestore, Sign-In and Secure Gemini Calls, Then Publishes to Your Personal Google Play Library; Web Apps Get Custom Domains

Google AI Studio product lead Logan Kilpatrick announced a batch of updates on X on November 17, alongside a Google Developers Blog post. The AI Studio Build agent can now use Gemini 4 Carbon and Hydrogen to turn a single prompt into a full-stack Kotlin and Jetpack Compose Android app, provisioning Cloud Firestore and Google Sign-In and calling Gemini through Firebase AI Logic and App Check so no API key ships in the APK. Finished apps can be published to your personal Google Play library and installed on your phone (US only for now). Web apps on paid Google Cloud projects can buy and connect a custom domain directly.

⚡ Key Takeaways
  • •Full-stack Android: choose Android in AI Studio Build and the Build agent generates a Kotlin and Jetpack Compose project, provisioning Cloud Firestore with Security Rules and Google Sign-In so data syncs across devices
  • •Secure Gemini calls: apps call Gemini through Firebase AI Logic and Firebase App Check, so no API key is embedded in the APK
  • •Personal Google Play publishing: set up a Play account from the publish panel with a standard Google account; AI Studio builds and signs the .aab and publishes it to your personal Play library for install on your phone; Logan Kilpatrick says this is US only for now
  • •Developer experience: a faster browser emulator with HD video and audio, one-click emulator screenshots attached to prompts for UI edits, custom fonts, and builds backed by the official Android CLI
  • •Custom domains: on paid Google Cloud projects, buy and connect a domain from the publish panel, with Cloud DNS and a managed SSL certificate set up automatically; the free ai.studio subdomain stays
Read details→
M
Microsoft AI@MicrosoftAI·Collected 20h ago
🚀 Release

Microsoft Launches MAI Image 2.7 and 2.7 Flash: #1 on Artificial Analysis Image Editing, Beating GPT Image 2.6 Sunburst (max) at About One-Fifth the Cost per Microsoft; Public Preview in Foundry at About $35 and $10 per 1,000 Images, Plus a New Canvas Workspace

Microsoft AI (MAI) released the image generation and editing models MAI Image 2.7 and MAI Image 2.7 Flash during Ignite on November 17. Microsoft says 2.7 ranks first on Artificial Analysis image editing (AA-Image-Editing v2.1) at 1,215 Elo, 4 points ahead of GPT Image 2.6 Sunburst (max), at about one-fifth of the cost. Both models are in public preview in Microsoft Foundry at about $35 per 1,000 images for 2.7 and about $10 for Flash, and are also offered through partners such as fal, OpenRouter and Vercel. MAI Labs also launched Canvas, an interactive editing workspace powered by Flash.

Microsoft Launches MAI Image 2.7 and 2.7 Flash: #1 on Artificial Analysis Image Editing, Beating GPT Image 2.6 Sunburst (max) at About One-Fifth the Cost per Microsoft; Public Preview in Foundry at About $35 and $10 per 1,000 Images, Plus a New Canvas Workspace
⚡ Key Takeaways
  • •Benchmarks: as of November 16, MAI Image 2.7 ranks first on AA-Image-Editing v2.1 at 1,215 Elo across 5,478 votes, 4 Elo ahead of GPT Image 2.6 Sunburst (max); on Arena, MAI is a top-two lab for image generation and single-image editing
  • •Pricing: public preview in Microsoft Foundry at about $35 per 1,000 images for MAI Image 2.7 and about $10 per 1,000 for 2.7 Flash; Microsoft puts the cost at roughly one-fifth of GPT Image 2.6 Sunburst (max)
  • •Editing: change an object's color, material or position without disturbing the rest of the image, with consistency across successive edits; both models support transparent backgrounds for product shots, logos and icons
  • •Canvas: a new interactive space in MAI Labs powered by 2.7 Flash; start from a prompt or sketch and select any part of an image to replace, move or restyle
  • •Availability: Foundry plus fal, Genspark, OpenRouter and Vercel; MAI Image is already the default model family in Bing Image Creator, and Figma Weave and Manus are among the platforms adopting it
Read details→
S
Satya Nadella@satyanadella·Collected 21h ago
🛠️ Tooling

Microsoft Ignite 2026 Agent Lineup: Copilot Code for Natural-Language Apps, the Long-Running Autopilot Agent and Multiplayer Copilot Workspaces Reach Frontier; Microsoft IQ Goes GA and Copilot SDK 2.0 Moves to a Rust Runtime

After its Ignite 2026 keynote on November 17, Microsoft published the Book of News, and Satya Nadella listed the highlights on X. Copilot Code uses the same agentic technology as GitHub Copilot to let anyone build dashboards and apps in natural language, with projects running in a sandbox inside the tenant; it reaches general availability in December. Autopilot is Microsoft's first enterprise-grade, long-running personal productivity agent with its own identity and memory, and Copilot Workspaces gives teams and Copilot a shared space. For developers, Microsoft IQ is generally available, Copilot SDK 2.0 ships a native Rust runtime, and Microsoft-Decision-1.1, built on Gemma 4, is generally available.

Microsoft Ignite 2026 Agent Lineup: Copilot Code for Natural-Language Apps, the Long-Running Autopilot Agent and Multiplayer Copilot Workspaces Reach Frontier; Microsoft IQ Goes GA and Copilot SDK 2.0 Moves to a Rust Runtime
⚡ Key Takeaways
  • •Copilot Code: build dashboards, trackers and apps in natural language; cloud projects run in a secure sandbox inside the tenant, and a local sandbox on Windows limits what it can read, write and reach on the network; available in the Frontier program on Windows, web, iOS and Android, with general availability in December
  • •Autopilot: Microsoft's first enterprise-grade, long-running personal productivity agent, with its own identity, memory and tool access in the tenant; it follows up on threads, runs recurring work and can be @mentioned in Teams; rolling out through the Frontier program
  • •Copilot Workspaces: the evolution of Copilot Notebooks, putting teams, shared documents and Copilot in one space where Copilot surfaces changes and open actions; bringing a personal Autopilot into a workspace is coming soon
  • •Microsoft IQ is generally available: it delivers enterprise context from Work IQ, Fabric IQ, Foundry IQ and Web IQ to Microsoft Copilot, GitHub Copilot, Foundry and any agent harness via MCP with no new subscription; Foundry Agent Optimizer is GA, and Programmed Training in preview lets engineers write their own SFT or RL loops on supported open-weight models
  • •Copilot SDK 2.0 is GA with the same agent harness Microsoft and GitHub run in production, a native Rust runtime and support for six languages; Microsoft-Decision-1.1 is GA, built on Gemma 4 26B A4B for routing and classification decisions, and available in Foundry, OpenRouter and Vercel
Read details→
G
GitHub@github·Collected 22h ago
🛠️ Tooling

GitHub Copilot Ignite Updates: Browser Use Enters Public Preview With a New Chrome Extension to Fill Forms, Pull Data and Debug Web Apps in Your Signed-In Browser; the Copilot App Adds Azure DevOps Repos and Code Review for Azure Repos Goes GA

During Microsoft Ignite on November 17, GitHub shipped a batch of Copilot updates. Paired with the new GitHub Copilot extension for Chrome, browser use is in public preview in Copilot CLI, the Copilot app, VS Code and JetBrains, so agents can fill out forms, read pages and test or debug the web app you are building in your signed-in Chrome or Edge. It runs in an isolated tab group and is off by default. The Copilot app adds Azure DevOps integration in public preview: clone a repo, have an agent branch, change code, open a pull request and address review comments. GHES support is now generally available. Copilot code review adds Adaptive, which picks review depth by pull request risk, plus secret detection, and code review for Azure Repos is now generally available.

GitHub Copilot Ignite Updates: Browser Use Enters Public Preview With a New Chrome Extension to Fill Forms, Pull Data and Debug Web Apps in Your Signed-In Browser; the Copilot App Adds Azure DevOps Repos and Code Review for Azure Repos Goes GA
⚡ Key Takeaways
  • •Browser use (public preview): with the GitHub Copilot extension for Chrome installed, Copilot CLI (enable via /plugin), the Copilot app, VS Code and JetBrains can drive Chromium browsers such as Chrome and Edge in an isolated tab group; it is off by default and permission levels live in the extension settings
  • •Copilot app plus Azure DevOps (public preview): clone an Azure DevOps repo, have an agent branch, change code and open a pull request, track it in the sidebar, click Fix on a review comment, or turn on agent merge to fix CI and complete the pull request once policies pass; cloud sessions are not available for Azure DevOps repos
  • •GHES: GitHub Enterprise Server support in the Copilot app is now generally available in version 1.2.0 or later
  • •Code review: Adaptive picks Lite, Balanced or Max per pull request based on complexity and risk, and cross-repository context and secret detection arrive in public preview
  • •Copilot code review for Azure Repos is generally available: turn it on per organization, project or repository, request reviews automatically via branch policy, and pay in GitHub AI credits billed to the linked Azure subscription
Read details→
G
GitHub@github·Collected 23h ago
📊 Benchmark

GitHub Copilot Fusion Tops the Public Terminal-Bench 4.1 Leaderboard: Fusion Max Resolves 71.2%, 5.4 Points Ahead of Claude Opus 5.5 and Fable 5.2 at 57% and 88% Lower Cost, and Fusion Comes to Web, Mobile and Visual Studio

GitHub updated Copilot Fusion, its runtime router that picks and orchestrates models for each task, on November 17. Fusion is now listed on the public Terminal-Bench 4.1 leaderboard hosted by Stanford, Harbor and the Laude Institute, where Fusion Max resolves 71.2% of tasks, 5.4 points ahead of Claude Opus 5.5 and Fable 5.2 in Claude Code, which tie for second, at 57% and 88% lower cost. Fusion can now also be selected on github.com, in Visual Studio and in the GitHub Mobile beta apps. The model pool changed too: Claude Sonnet 5.5 joins Fusion Max, GPT-6.1 Terra joins Fusion High, and Microsoft's new MAI-Code-2 joins Fusion Standard.

GitHub Copilot Fusion Tops the Public Terminal-Bench 4.1 Leaderboard: Fusion Max Resolves 71.2%, 5.4 Points Ahead of Claude Opus 5.5 and Fable 5.2 at 57% and 88% Lower Cost, and Fusion Comes to Web, Mobile and Visual Studio
⚡ Key Takeaways
  • •Public Terminal-Bench 4.1 leaderboard: Copilot Fusion (max) is first at 71.2% plus or minus 2.9%, Opus 5.5 (max) and Fable 5.2 (max) tie for second at 65.8%, and Gemini 4 Carbon (high) is fourth at 64.5%
  • •Cost: the leaderboard shows Fusion Max used 3.3B tokens and about $2.0k for the full run, versus about $4.7k for Opus 5.5 and about $16.5k for Fable 5.2
  • •How it works: Fusion is not a single model but a runtime router that picks one of three execution patterns per prompt: Single, Cascade (a cheaper model drafts and a judge decides whether to escalate) or Critique (several models critique in parallel before a revision)
  • •New surfaces: Fusion can be selected on github.com, in Visual Studio and in the GitHub Mobile beta apps for iOS and Android, with production mobile apps coming shortly
  • •Model pool: Claude Sonnet 5.5 joins Fusion Max, GPT-6.1 Terra joins Fusion High and MAI-Code-2 joins Fusion Standard; usage is billed at each model's standard rate, with a 30% model cost discount for paid plans through November 30
Read details→
M
Microsoft@Microsoft·Collected 23h ago
🚀 Release

Microsoft Launches MAI Code 2 and MAI Code 2 Flash at Ignite, Now in GitHub Copilot: Code 2 Reaches 98% of GPT-6.1 Sol at About 40% of the Price, Alongside the MAI Cyber 2 Flash Security Model

Microsoft AI (MAI) announced two coding models at Ignite on November 17. MAI Code 2 targets multi-file implementation, deep debugging and repo-wide reasoning, while MAI Code 2 Flash handles high-volume work such as code explanations, targeted edits and routine fixes. Both are available in GitHub Copilot today, Code 2 Flash is in limited-access preview in Microsoft Foundry, and the VS Code team says both are 50% off through May 17, 2027. Microsoft says that averaged over four everyday enterprise coding benchmarks, Code 2 reaches 98% of GPT-6.1 Sol at about 40% of the price and 96% of Claude Opus 5.5 at 5% of the price. It also released the MAI Cyber 2 Flash security model.

Microsoft Launches MAI Code 2 and MAI Code 2 Flash at Ignite, Now in GitHub Copilot: Code 2 Reaches 98% of GPT-6.1 Sol at About 40% of the Price, Alongside the MAI Cyber 2 Flash Security Model
⚡ Key Takeaways
  • •Two models: MAI Code 2 for taking code to production, deep debugging, multi-file implementation, repo-wide reasoning and UI; MAI Code 2 Flash for code explanations, targeted edits, routine fixes and rapid iteration
  • •Official scores: averaged over DeepSWE 1.1, Terminal Bench 2.1, SWE-Atlas and FrontierCode v1.1, Code 2 reaches 98% of GPT-6.1 Sol at about 40% of the price and 96% of Claude Opus 5.5 at 5% of the price, with cost measured at the 50% promotional price
  • •Code 2 Flash matches GPT-6 Luna and Claude Haiku 5.5 on three everyday coding benchmarks at 65% and 20% of their price; in blind frontend app reviews, raters preferred Code 2 over GPT-6 Sol 67% of the time
  • •Availability: in GitHub Copilot today, including VS Code; Code 2 Flash is in limited-access preview in Microsoft Foundry, with OpenRouter, Vercel, OpenCode and Kilo coming soon
  • •MAI Cyber 2 Flash: 92.3% on CyberGym-E2E versus 91.5% for GPT-6 Astra at under 2% of the cost per task, with a 1 million token context; it comes only to Microsoft security products such as MDASH from mid-December
Read details→
O
OpenAI@OpenAI·Collected 1d ago
🚀 Release

OpenAI Releases GPT-6.2 Astra and GPT-6.2 Sol: Artificial Analysis Intelligence Index 62 Places Astra Second Behind Claude Fable 5.2 at 65; Astra Cuts Prices 20% to $8/$40 per Million Tokens

OpenAI released GPT-6.2 Astra and GPT-6.2 Sol on November 17 (U.S. time). Starting today they are available to Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex, and developers can call gpt-6.2-astra and gpt-6.2-sol in the API, with Ultrafast available for both at launch. Astra API prices are 20% below GPT-6 Astra: $8 input, $0.40 cached input and $40 output per million tokens. Sol keeps GPT-6.1 Sol pricing of $2, $0.10 and $10. Independent evaluator Artificial Analysis scores Astra 62 on the Intelligence Index (Max), second behind Claude Fable 5.2 at 65; Sol scores 57, matching GPT-6 Astra at about 15% of the cost per task. On the Coding Agent Index, Artificial Analysis finds Astra flat versus GPT-6 Astra and Sol three points lower than GPT-6.1 Sol.

OpenAI Releases GPT-6.2 Astra and GPT-6.2 Sol: Artificial Analysis Intelligence Index 62 Places Astra Second Behind Claude Fable 5.2 at 65; Astra Cuts Prices 20% to $8/$40 per Million Tokens
⚡ Key Takeaways
  • •Artificial Analysis Intelligence Index (Max): GPT-6.2 Astra scores 62 to rank second, behind Claude Fable 5.2 at 65, five points above GPT-6 Astra and ahead of Gemini 4 Carbon (60) and Claude Opus 5.5 (59); Sol scores 57, matching GPT-6 Astra at about 15% of the cost per task
  • •OpenAI figures on Agents' Last Exam: Astra is 5.2 points above GPT-6 Astra at under a third of the cost and 5.5 points above Claude Fable 5.2 at 5% of the cost
  • •Science, 3D and robotics: Artificial Analysis says Astra leads Terminal-Bench-Science 0.1; on AA-3D Astra scores 1318, ahead of Fable 5.2 at 1271 and Sol at 1259; on AA-Robotics v0.1 the scores are 79, 78 and 75
  • •Coding: on OpenAI-cited FrontierCode 2.0, Sol lands within 0.5 points of Astra at about one fifth the cost; on the AA Coding Agent Index (Codex harness, max) Astra is flat versus its predecessor and Sol is 3 points lower, driven by a 14-point drop on SWE-Atlas-QnA-AA
  • •Pricing and safety: Astra is $8 input, $0.40 cached input and $40 output per million tokens (20% below GPT-6 Astra); Sol stays at $2, $0.10 and $10. Astra is the first externally released model rated High for AI Self-improvement
Read details→
G
Google Gemini@GeminiApp·Collected 1d ago
🚀 Release

Google Makes Tasks in Gemini Free for Hundreds of Millions of Users in Six Countries on Gemini 4 Hydrogen, Adding Phone Calls on Your Behalf, AP2 Checkout With Google Pay and Email Forwarding in the U.S.

On November 17 Google announced that Tasks in the Gemini app is now free for hundreds of millions of eligible users in the U.S., Japan, Korea, Canada, Australia and India, running on its new Gemini 4 Hydrogen model. Chat stays for quick answers, while Tasks handles longer projects and recurring schedules in the background. A set of advanced features is rolling out first in the U.S.: Gemini can place phone calls for you, including calling several shops at once for quotes or bookings; it can check out with Google Pay through the Agent Payments Protocol (AP2) under a spending cap you set; and you can hand it work by forwarding email to [email protected] or by text message. Google did not give a date for bringing the advanced features to other countries.

Google Makes Tasks in Gemini Free for Hundreds of Millions of Users in Six Countries on Gemini 4 Hydrogen, Adding Phone Calls on Your Behalf, AP2 Checkout With Google Pay and Email Forwarding in the U.S.
⚡ Key Takeaways
  • •Who gets it free: hundreds of millions of eligible users in the U.S., Japan, Korea, Canada, Australia and India; flip the home screen toggle from Chat to Task in the Gemini app
  • •Model: Tasks now runs on Gemini 4 Hydrogen, which Google says follows instructions more closely and picks the right tools for the job
  • •U.S.-first advanced features: phone calls on your behalf, including several businesses at once, cancelable mid-call, with a business opt-out; Google Pay checkout through AP2 with a hard spending cap and confirmation before buying by default
  • •New entry points: forward email to [email protected] or text Gemini via Google Messages or iMessage; it can also edit Docs and Sheets, send Gmail and schedule Calendar events, and connects to more than 60 apps
  • •Security: each task runs in a hardware-isolated virtual machine dedicated to your account on Google Cloud Agent Compute Engine, plus prompt-injection filters and Safe Browsing
Read details→
D
Demis Hassabis@demishassabis·Collected 1d ago
🚀 Release

Isomorphic Labs Raises $5 Billion Series C Led by Greenoaks and Sequoia, With a16z and Alphabet, to Scale Its IsoDDE AI Drug Design Engine and Move First Drugs Toward Clinical Trials

Isomorphic Labs, the AI drug discovery company spun out of Google DeepMind, announced a $5 billion Series C on November 17. Greenoaks and Sequoia Capital co-led the round, with a16z and other new investors, continued significant investment from Alphabet, and participation from all existing investors including Thrive Capital, MGX, GV, CapitalG, Temasek and the UK Sovereign AI Fund. The money will mature its in-house drug design engine IsoDDE, push into new biology, build global AI-led bioengineering infrastructure and move oncology and immunology candidates toward clinical trials. CEO Demis Hassabis announced it himself on X. No valuation was disclosed.

Isomorphic Labs Raises $5 Billion Series C Led by Greenoaks and Sequoia, With a16z and Alphabet, to Scale Its IsoDDE AI Drug Design Engine and Move First Drugs Toward Clinical Trials
⚡ Key Takeaways
  • •Size and investors: a $5 billion Series C co-led by Greenoaks and Sequoia Capital, with a16z and other new investors, continued significant investment from Alphabet and all existing investors; the round was oversubscribed and no valuation was disclosed
  • •Core asset IsoDDE: it unifies a suite of foundational simulation AI models that work across therapeutic areas and drug modalities, combined with proprietary agents that reason over complex biological systems to design molecules
  • •Progress: the company says internal oncology and immunology programs and partnered programs are showing positive preclinical results, setting the stage for clinical entry
  • •Use of funds: maturing IsoDDE, building a partner ecosystem for mechanistic disease biology, global AI-led bioengineering infrastructure, and AI-first translational and clinical development capacity
  • •Demis Hassabis wrote on X that the mission is to solve all disease by reimagining drug design from first principles with AI and radically cutting the time and cost to find new medicines
Read details→
P
Perplexity Developers@perplexitydevs·Collected 1d ago
🛠️ Tooling

Perplexity Launches Monitor API: Scheduled Searches on a Topic or Text-Change Tracking for 1 to 10 Pages, With Webhook Alerts, at $1 per 1,000 Runs or Fetched URLs

Perplexity's developer account released the Monitor API on November 16. It runs checks on the web on a schedule and saves the results, which you can read through the API or receive as webhook notifications. There are two types: Search runs a query on a schedule and keeps only new, relevant results, while Page tracks 1 to 10 known URLs, sets a baseline on the first fetch and then returns meaningful text changes. The official pricing page lists $0.001 per succeeded Search run or per counted fetched URL in a succeeded Page run, which is $1 per 1,000, and failed or skipped runs are free. It is supported in the Python and TypeScript SDKs and the API Console.

⚡ Key Takeaways
  • •Two monitor types: Search (runs a query on a schedule and keeps new, relevant results) and Page (tracks text changes at 1 to 10 known URLs); one monitor cannot mix the two
  • •Flow: the first run starts asynchronously right after creation, later runs follow the schedule (for example weekly) until you pause or delete the monitor; an optional goal describes the results you want
  • •Each run stores kept findings in changes and fetch errors in target_errors; read them by polling the API or configure a public HTTPS webhook for notifications
  • •Pricing: $0.001 per succeeded Search run and $0.001 per counted fetched URL in a Page run, or $1 per 1,000; failed and skipped runs are free
  • •Perplexity CEO Aravind Srinivas promoted it on X as a way to track any event, topic or webpage change and get webhook updates
Read details→
P
Perplexity@perplexity_ai·Collected 1d ago
🚀 Release

Perplexity Launches Frontier Research: More Than 10,000 GPUs Behind an "Automated Scientist" Effort Focused on Multi-Agent Orchestration, Pre-Training, Continual Learning and RL, With Papers, Open Weights, Data and Evals Promised

On November 16 Perplexity announced Perplexity Frontier Research on its official X account and a new site, research.perplexity.ai, with the goal of building a system that can run the research loop on its own. The official page names Andrew Gordon Wilson as research lead, says the effort is backed by more than 10,000 GPUs, lists six areas (multi-agent orchestration, pre-training, continual learning, reinforcement learning, systems and frontier evaluations) and commits to publishing papers, models, data and benchmarks. CEO Aravind Srinivas said on X that Perplexity will open-source its models and tooling. The team is hiring in San Francisco, Palo Alto, New York and London, with a December 5 priority deadline.

⚡ Key Takeaways
  • •Goal: an "automated scientist", a multi-agent system that asks questions, searches wide and deep, forms hypotheses, runs experiments and keeps iterating to produce interpretable scientific understanding rather than only black-box predictions
  • •Resources: the official page says the effort is backed by more than 10,000 GPUs; Srinivas said on X at least 10,000
  • •Six areas: multi-agent orchestration, pre-training (open weights, new architectures, synthetic data), continual learning (curiosity, memory, persistence), reinforcement learning, systems (inference, training, runtimes, security) and frontier evaluations
  • •The multi-agent orchestration goal is the best-quality output for whatever cost the user wants to spend, for example a small on-device model that asks a larger model for help only when it helps
  • •Openness: papers, models, data and benchmarks will be published; the page already lists evals such as WANDR and DRACO and open weights such as Decider v1.2 (27B); December 5 is the priority application deadline
Read details→