zzir retweeted
1/ Any model on OpenRouter can now run code in a hosted Linux container.
Add one tool to a Responses API request. The model writes the script, runs it, and reads back the result. Nothing to install, nothing to host.
Meet openrouter:shell.
To get started:
"tools": [{ "type": "openrouter:shell", "parameters": { "engine": "openrouter" } }]
zzir retweeted
At 250 billion DNS cache entries, one wasted byte costs 250 GB of RAM.
Five Rust optimizations later: 100 TB freed, inserts 43% faster, lookups 19% faster.
We didn't trade speed for space. cfl.re/4gyL9Jj
zzir retweeted
Today, @ducklabs_com is joining @awscloud. The move is expected to be completed by early September.
Joining AWS gives DuckLabs the resources and reach to bring DuckDB, DuckLake, and the Quack protocol to many more developers and organizations – and to pursue ideas at a scale that would have been difficult for us to reach alone.
The DuckLabs team will remain together in Amsterdam and will continue working for the Duck Stack community as an AWS subsidiary.
Most importantly, #DuckDB and the other open source components of the Duck Stack will remain free and open source under the MIT license, with the non-profit DuckDB Foundation continuing its stewardship of the projects.
This is a significant moment for DuckLabs and the Duck Stack community. It marks the end of one chapter that we are immensely proud of, and the beginning of another that we believe will take DuckDB much further.
Read the full announcement here: ducklabs.com/ducklabs-is-joi…
#AWS
zzir retweeted
This Tuesday, we announced Quack, our new protocol that turns DuckDB into a client-server database. Watch Hannes' talk, recorded at the AI Council, where he explained how Quack works in practice, what stack it's built on, how it performs, and what our long-term ambitions are.
zzir retweeted
Today, Cloudflare Email Service enters public beta with the infrastructure layer to make that easy: send, receive, and process email natively from your agents. cfl.re/41H3uga
zzir retweeted
We are happy to release DuckLake v1.0, a production-ready lakehouse format specification. Its reference implementation, the ducklake DuckDB extension, is available as of today in DuckDB v1.5.2.
Since releasing the DuckLake manifesto in May 2025, we have seen massive adoption, with DuckLake deployed in production at multiple organizations, third-party clients supporting DuckLake, and even an upcoming O'Reilly book. DuckLake v1.0 ships many new features (inlining, partitioning, bucketing, type system improvements) and guarantees backwards-compatibility in the specification.
zzir retweeted
Google releases Gemma 4. ✨
Gemma 4 introduces 4 models: E2B, E4B, 26B-A4B, 31B.
The multimodal reasoning models are under Apache 2.0.
Run E2B and E4B on ~6GB RAM, and on phones.
Run 26B-A4B and 31B on ~18GB.
GGUFs: huggingface.co/collections/u…
Guide: unsloth.ai/docs/models/gemma…
zzir retweeted
Meet Gemma 4!
Purpose-built for advanced reasoning and agentic workflows on the hardware you own, and released under an Apache 2.0 license.
We listened to invaluable community feedback in developing these models. Here is what makes Gemma 4 our most capable open models yet: 👇
zzir retweeted
DuckLake 1.0 will ship with the “data inlining” feature: small updates are staged in the catalog instead of creating many small files.
In today's blog post, Pedro Holanda shows how DuckLake implements inlining and shows an almost 1000× speedup on certain operations.
ducklake.select/2026/04/02/d…
zzir retweeted
Introducing the new /crawl endpoint - one API call and an entire site crawled.
No scripts. No browser management. Just the content in HTML, Markdown, or JSON.
zzir retweeted
Qwen releases 4 new Qwen3.5 Small models!
Qwen3.5: 0.8B • 2B • 4B • 9B
Run Qwen3.5-0.8B, 2B and 4B on your phone. Run 9B on 6GB RAM.
The vision reasoning LLMs perform better than models 4x their size.
GGUFs: huggingface.co/collections/u…
Guide: unsloth.ai/docs/models/qwen3…
🚀 Introducing the Qwen 3.5 Small Model Series
Qwen3.5-0.8B · Qwen3.5-2B · Qwen3.5-4B · Qwen3.5-9B
✨ More intelligence, less compute.
These small models are built on the same Qwen3.5 foundation — native multimodal, improved architecture, scaled RL:
• 0.8B / 2B → tiny, fast, great for edge device
• 4B → a surprisingly strong multimodal base for lightweight agents
• 9B → compact, but already closing the gap with much larger models
And yes — we’re also releasing the Base models as well.
We hope this better supports research, experimentation, and real-world industrial innovation.
Hugging Face: huggingface.co/collections/Q…
ModelScope: modelscope.cn/collections/Qw…
zzir retweeted
Let’s dive deeper into WebMCP's new APIs. First, the Declarative API: it's built for high-precision, form-based tasks. By defining tools in your HTML, you enable agents to book flights or file support tickets with speed and reliability that scraping can't match 🤖🧵 (1/3)
zzir retweeted
📣 @OpenAIDevs GPT-5.3-Codex is now generally available and rolling out in GitHub Copilot.
Early testing shows
➡️ It reaches new high scores on benchmarks we use to evaluate coding, agentic, and real-world capabilities
➡️ 25% faster performance than GPT-5.2-Codex on agentic coding tasks
➡️ Improved reasoning and execution in complex workflows
Try it out in @code or Copilot CLI.
github.blog/changelog/2026-0…
zzir retweeted
Officially launching X API Pay-Per-Use
The core of X developers are indie builders, early stage products, startups, and hobbyists
It’s time to open up our X API ecosystem and instill a new wave of next generation X apps
We’re so back. developer.x.com
zzir retweeted
Alright - here's my most powerful vibe coding prompt ever.
Plug this straight into the new Opus 4.6 model, and you'll literally be able to ship anything.
SEO optimised websites, fully functional Claude apps, personal dashboards - literally anything.
Takes <5 minutes:
zzir retweeted
GPT-5.3-Codex is here!
*Best coding performance (57% SWE-Bench Pro, 76% TerminalBench 2.0, 64% OSWorld).
*Mid-task steerability and live updates during tasks.
*Faster! Less than half the tokens of 5.2-Codex for same tasks, and >25% faster per token!
*Good computer use.
zzir retweeted
🚀 Introducing Qwen3-Coder-Next, an open-weight LM built for coding agents & local development.
What’s new:
🤖 Scaling agentic training: 800K verifiable tasks + executable envs
📈 Efficiency–Performance Tradeoff: achieves strong results on SWE-Bench Pro with 80B total params and 3B active
✨ Supports OpenClaw, Qwen Code, Claude Code, web dev, browser use, Cline, etc
🤗 Hugging Face: huggingface.co/collections/Q…
🤖 ModelScope: modelscope.cn/collections/Qw…
📝 Blog: qwen.ai/blog?id=qwen3-coder-…
📄 Tech report: github.com/QwenLM/Qwen3-Code…
