Search: “models”

This post documents a Wordle game result where the player solved puzzle #1,762 in four attempts, using the standard color-coded feedback system (gray for incorrect letters, yellow for correct letters

You can also set your autocompact threshold yourself and effectively lower your context window if you'd prefer. For example, 400k context is a good compromise: CLAUDE_CODE_AUTO_COMPACT_WINDOW=400000 c

Cristobal Valenzuela, co-founder and CEO of Runway, posted a cryptic and anticipatory message on X suggesting that something significant is imminent or that current expectations about 2026 may be prem

ARC-AGI-3 has the lowest human bar of any AI benchmark out there. Almost all benchmarks require specialized knowledge that make them inaccessible to 99%+ of humans (like, say SWE-Bench). ARC-AGI-3 is

Been waiting a month for Anthropic to answer a simple usage question about Claude Code subscriptions Have I been ghosted Can I get some questions answered by someone at Anthropic? 1. Can you use an OA

btw the famous slack chart is slack propaganda and everyone who cites it is legally obligated to also link to @sophiebits Every time I see a tweet saying “I can vibe code this in a weekend” - I think

Budget blown on closed APIs is a solvable problem. Simply replace 20m of closed-source tokens with 1m on Minimax M2.7. Frontier performance. 20x lower cost. No rate limits. If you're rethinking your A

Comparing hardware on InferenceX just got a massive upgrade. Our new changelog update lets you compare every change for any gpu all on the same chart. Just select any model and GPU config, then select

`deepagents deploy` now supports user scoped memory! add a user/ directory in your project so each user gets their own writable AGENTS.md, seeded on first deploy and persisted across conversations. yo

Document OCR benchmarks are still an open problem Existing document OCR benchmarks are either too narrowly focused on a specific type (e.g. FinTabNet, ChartQA), or on documents that aren’t reflective

Get Started in ComfyUI 1. Update ComfyUI to 0.19.1 (Desktop and Cloud will be available soon) 2. Go to Template → Search for ERNIE-Image Select the ERNIE-Image workflow 3. Download the missing models

I ran the same prompt for a London Estuary accent, a Newcastle accent and an Exeter, Devon accent - all three audio files are now embedded in my blog post https://simonwillison.net/2026/Apr/15/gemini-

If it can’t parse, it can’t perform. ❌ Reliable document understanding is fundamental to enterprise-grade #AgenticAutomation. We're excited to see the release of ParseBench, a new open-source benchmar

It’s finally here! 🚀 Huge thanks to the @opencode team for the seamless integration. Qwen3.6-Plus and Qwen3.5-Plus are now live in Go! Update now to try it out! 👇 Qwen3.6 Plus and 3.5 Plus now availab

Just a Hermes Agent, a skill, and a dream The crazy part? This was done (nearly) fully autonomously! Only 8 prompts from the human in the loop. Just a Hermes agent, a skill, and a dream. 🐉 I told my A

Let's talk parsing tables. Two days ago we launched ParseBench,the first document OCR benchmark built for AI agents. This deep dive breaks down TableRecordMatch (GTRM), our metric for evaluating compl

NASA wasn’t “closed” before Trump. The US doesn’t pay NATO anywhere close to “trillions of dollars” or “hundreds of billions of dollars a year.” Trump hasn’t built anywhere close to “over 1,000 miles

Parsing complex tables in PDFs is extremely challenging. Existing metrics for measuring table accuracy, like TEDS (tree edit distance similarity), overweight exact table structure and underweight sema

SpaceX is moving at literal light speed right now SpaceX launched 2 more Falcon 9s in under 24 hours AGAIN - Florida in the morning, California at night Even with unlimited budgets, government space p

Starlink, SpaceX's satellite internet service, has been praised for delivering superior connectivity compared to competing options, including in-flight internet services on commercial aircraft. The se

Tesla has just released a new app update. Clips downloaded to your phone now finally include details such as speed, steering wheel angle, and Self-Driving state. No more screen recording FSD clips! He

This is becoming a pattern in AI that makes talking about capabilities challenging. First, there are overstated claims (like the flubbed Erdos problems last year), then minor wins (AI helps with disco

Weird how some people always target open-source in AI! First it was: “Open-source AI will destroy the world” (spoiler: it didn't and it won't) Now: “Open-source is a cybersecurity threat because of AI

This post from Anthropic's X (formerly Twitter) account shows a Wordle game result for puzzle number 1,760, solved in 4 out of 6 attempts. The emoji grid displays the progression of guesses, with blac

Clem Delangue (or someone in their network) shared excitement about successfully running GLM 5.1 locally, noting it appears to be functional and effective. The post highlights the cost advantage of ru

Claude Code Routines are here! In addition to a schedule, you can now trigger templated agents via GitHub event or API – with our infra & your MCP+repos They've changed how we do docs, backlog mainten

Every technological revolution created as many jobs as it destroyed, because we find that there is always more work to do. Things that were expensive or impossible become cheap, and we are allowed to

Excited to be part of @ollama Gemma Day tomorrow in Palo Alto! Khoa Pham (@kwafam7) from @radixark will demo Gemma 4 in production with SGLang 🚀 Come say hi! 📷 RSVP: https://luma.com/ollama-gemma4 Oll

SpaceX's Falcon 9 rocket successfully launched 29 Starlink satellites from a Florida launch site, further expanding SpaceX's broadband internet constellation. This mission is part of SpaceX's ongoing

ICYMI -- last week we released `deepagents deploy`, the fastest way to take a highly capable, long running agent to production. agents are becoming more and more standardized, and we're betting on thi

It looks like everyone is finally catching up with the fact that agent sessions in CLI mode can only get you so far. It makes sense that the new Codex app, Cursor, and Claude Code (desktop) feel and l

llama.cpp built from source + qwen3.5 27b running locally + hermes agent on top + camofox for scraping (tls spoofing) + scweet to scrape x (no api keys) + tailscale to access from other devices bro Me

LLM Knowledge Base → Slides When @karpathy shared his LLM Knowledge Base setup, many were wondering how to generate more visual forms of the wiki. There are many options, but I think @GammaApp is one

New research result: we use Claude to make fully autonomous progress on scalable oversight research, as measured by performance gap recovered (PGR). Claude iterates on a number of different techniques

Now in research preview: routines in Claude Code. Configure a routine once (a prompt, a repo, and your connectors), and it can run on a schedule, from an API call, or in response to an event. Routines

Ollama and Google Gemma team is hosting an Ollama Gemma Day in Palo Alto tomorrow night (Wednesday, April 15th at 6pm). Lots of amazing speakers from @GoogleDeepMind and @sgl_project / @radixark! Look

ParseBench is the most comprehensive OCR benchmark for real-world enterprise documents: financial filings, contracts, insurance documents, and more. We evaluate across 5 dimensions that are present am

Soon, at each gradual improvement level of AI, you will start to see large discrete jumps in ability in economically important areas, because the previous AI ability level in some aspect of the job bo

This is why we released liteparse :) Free, open-source, designed for agents. Natively supports OCR / screenshotting for deeper visual understanding in a document when needed. @kepano I just tried it t

Today we're launching a rebuilt version of Claude Code on desktop. The app has been redesigned for the ground up to make it easier than ever to parallelize work with Claude. I haven't opened an IDE or

We’re expanding Trusted Access for Cyber with additional tiers for authenticated cybersecurity defenders. Customers in the highest tiers can request access to GPT-5.4-Cyber, a version of GPT-5.4 fine-

We've been working on this for a while. Can't wait to hear what you think We've redesigned Claude Code on desktop. You can now run multiple Claude sessions side by side from one window, with a new sid

Anthropic's official X (Twitter) account shared a Wordle result showing the puzzle was solved in 3 out of 6 attempts for Wordle #1,759. The grid reveals the solver had no correct letters in the first

Armageddon awaits the Senate GOP if we don’t get the SAVE America Act passed into law Let’s freaking go! McLaughlin poll released today shows 1 in 8 would-be GOP voters stay home in the midterms in ME

'Both the utopia and apocalypse scenarios ignore AI's limitations.' Prof. Yoav Shoham, one of the world's leading artificial intelligence researchers and founder of @AI21Labs, discusses hype cycles, u

ComfyUI has integrated support for Seedance 2.0, a video generation model, making it accessible directly within the ComfyUI workflow interface. The announcement directs users to the official ComfyUI b