Claude Fable 5 ranks #1 in Code Arena: Frontend, leading by a wide margin over Opus-4.8. Highlights: - #1 in every sub leaderboard: HTML, React - #1 in every sub category: Brand & Marketing, Reference-Based Design, Data & Analytics, Consumer Product, Gaming, Simulations, and Content Creation Tools. Huge congrats to @AnthropicAI for this milestone! The thread breaks down how Claude Fable 5 ranks across single-modality arenas.

Things you should do this summer in San Francisco NorCal instead of sitting inside with Claude Code: 🌁 Grab fresh oysters from Tomales Bay 🌁 Pick strawberries, cherries, & blackberries in Brentwood 🌁 Walk the SF Crosstown trail 🌁 Picnic in Dolores, GGP, Lafayette, Crissy Fields, or Alamo Square parks 🌁 Go on a wine tour in Sonoma & Napa 🌁 Take the ferry from SF to Sausalito 🌁 Drive down to Santa Cruz and grab burritos on the beach 🌁 Visit the Ferry Building or Fort Mason farmers markets 🌁 Take a trip to Muir Woods 🌁 Drive down Highway 1 for insane views 🌁 Explore or camp in Carmel 🌁 Hike Mission Peak in Fremont 🌁 Visit Yosemite/Halfdome 🌁 Golf in Half Moon Bay 🌁 Polar Plunge at Aquatic Park or Ocean Beach 🌁 Dine at the Taco Bell Cantina in Pacifica

Introducing Agent Arena: real-world agentic evals at scale. How do you evaluate agents doing actual work? We measure millions of live sessions where real users accomplish real tasks. On Arena, models now get web search, filesystem, and terminal tools to complete complex workflows: writing code, creating slide deck, researching the web, building apps, and analyzing documents. Every session produces rich signals. Users iterate with the agent turn-by-turn: approving, editing, correcting, praise or expressing frustration. The environment gives feedback too: shell errors, tool failures, recovery attempts, and more. Our leaderboard measures each model's agentic performance using causal inference across five signals: task success, steerability, error recovery, user praise vs. complaint, and tool hallucination. This leaderboard snapshot is built from 300K tasks, 2M tool calls, and 40M lines of code by agents. Top labs in Agent Arena: - #1 @OpenAI: GPT-5.5 (High) - #2 @AnthropicAI: Claude-Opus-4.7 (Thinking) - #3 @Zai_org: GLM-5.1 - #4 @GoogleDeepMind: Gemini-3.1-Pro - #5 @Kimi_Moonshot: Kimi-K2.6 More analysis in the thread, with the full technical blog below.

Opus4.8すごいすね…………

IMO a researcher studies a problem that may not be solvable, while an engineer solves a problem that is considered solvable.

Fun fact, GPT 5.5 is very good at Game Dev Game Dev is the notable category where @OpenAI consistently beats out @AnthropicAI's Claude models Upon code inspection, our @Designarena team found that GPT 5.5's frontend verbosity plays in its favor for game dev - it consistently created games with the most functional features Congrats to @OpenAI for establishing the new Game Dev frontier!

New preprint alert 🚨 Can LLM agents develop video games? We release GameDevBench, the first benchmark evaluating agentic game development in a game engine, Godot. We also present two simple multimodal feedback mechanisms that lead to immediate performance gains. /🧵

Introducing Moonlake's 3D Agent. Our agent acts like a technical artist that can build and reconstruct articulated assets and large-scale editable scenes with hundreds of objects from a single image and can improve its generations continuously. Learn more in the thread below.

Tired of evaluating LLMs on made-up problems that look nothing like real tasks? Introducing EDIT-Bench, a code editing benchmark built from in-the-wild user interactions in VSCode. Real-world edits are challenging: 𝗼𝗻𝗹𝘆 𝟭/𝟰𝟬 𝗺𝗼𝗱𝗲𝗹𝘀 𝘀𝗰𝗼𝗿𝗲 > 𝟲𝟬% 𝗽𝗮𝘀𝘀@𝟭.

Benchmarks? Where we’re going, we don’t need benchmarks.

Indie game Slay the Spire 2 has surpassed 500,000 concurrent players on Steam The rougelike is now in the top 20 games with highest all-time peaks on Valve's platform