Chris Fryer

.@opencode has a /tree now, and context is way easier to manage.

🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. 🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. 🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice. Try it now at chat.deepseek.com via Expert Mode / Instant Mode. API is updated & available today! 📄 Tech Report: huggingface.co/deepseek-ai/De… 🤗 Open Weights: huggingface.co/collections/de… 1/n

> Random Guy with no Biology Degree > zero Lab experience > asked Claude how to build a wet lab at Home > extracted DNA and sequenced his own genome > Using a $1,000 device: MinION > barrier has never been this lower Reddit users are way ahead of avg peoples.

McKinsey analyst in 2030:

Message from a very good software engineer: “i switched to codex because of anthropics shenanigans and discovered it's a better coding model. trying kimi k2.6 right now and i'm blown away by how good it is. i still have an anthropic plan but 99% sure i'm going to cancel it” what are you doing @AnthropicAI?

everyone clapping for dario being conservative with compute great example of how everyone is wrong about everything

OpenAI just released a new open-source model it's "a bidirectional token-classification model for personally identifiable information (PII) detection and masking in text" github.com/openai/privacy… huggingface.co/openai/privacy…

he's right. it's brutal to watch more spacex value get destroyed

I don't think LLMs will improve much from here

36,000 engineering students from IT programs were tested in India. 95% were determined to be unfit for actually working in the field. Only 1.4% could write functionally correct and efficient code. India has a massive issue with cheating and scams in education. Fake degrees

Linux kernel 7.1 introduces a new in-kernel NTFS driver, a complete ground-up rewrite that delivers native read and write access to Windows NTFS volumes directly within the kernel without any userspace tools. It replaces the slower FUSE-based NTFS-3G and the previous NTFS3 driver. Performance gains include 3-5 percent faster single-threaded writes, 35-110 percent faster multi-threaded writes, and mounting times for large drives up to four times quicker. The driver also features lower CPU overhead, better integration with modern kernel infrastructure, and improved reliability. This update marks a major step forward for Linux-Windows storage interoperability. Via:tomshardware

guys im beginning to think theres a bubble

BREAKING: Vibe-coding platform Lovable reportedly suffered a breach that exposed users’ AI chat histories, source code, & database credentials.

where i'm at on LLMs: 1. LLMs are NOT going to outmode software engineers, not now, not ever 2. LLMs alone are a dead end for attaining anything like AGI 3. forget about LLMs achieving ASI 4. LLMs will only make good engineers faster, marginally 5. LLMs will remain poor at architecture and design this is 100% because LLMs cannot reason and no, generating a bunch of hidden context is not "reasoning" as always, when the same people are making claims like this, have a look at what they stand to gain and to them, i warn directly: you are going to fall. hard

we've done instrumentation work in opencode recently so we can get traces out they get shipped to a local otel tui (by kit) which also has endpoints for the agent to read this data provides a good feedback loop for the agent to analyze issues - will do a video soon

I built this plugin for @opencode called Session Graph. It shows how all your chat branches connect. So when you fork a session, you can actually see where it came from, where it split, and how to get back. Makes branching chats way less confusing. @thdxr

Google DeepMind researcher argues that LLMs can never be conscious, not in 10 years or 100 years. "Expecting an algorithmic description to instantiate the quality it maps is like expecting the mathematical formula of gravity to physically exert weight."

and so are rocks (though, less conscious than bananas)

It looks like Anthropic changed claude code’s cache TTL from 1h to 5m in March, causing significant quota and cost inflation.

UPDATE: We were able to replicate the Mythos findings using existing models (GPT5.4) Writeup coming early next week, no BS prompts, it's real reproduction