Nick
@nickbaumann_
newspapers, catalogs, & almanacs @cline | product of @UWMadison 🦡
Something’s rotten in startup land. When did we become orthodox? We used to celebrate mavericks. Now we only accept founders and startups from Innovation™️ University. We used to be pirates—now we’re the Port Authority. Stop gatekeeping disruption and raise the black flag! 🏴☠️
Qwen3-Coder is now available in Cline 🧵 New 480B parameter model with 35B active parameters. > 256K context length natively > comparable performance to Claude Sonnet 4 > SoTA among open source models
>>> Qwen3-Coder is here! ✅ We’re releasing Qwen3-Coder-480B-A35B-Instruct, our most powerful open agentic code model to date. This 480B-parameter Mixture-of-Experts model (35B active) natively supports 256K context and scales to 1M context with extrapolation. It achieves…
Kimi K2 outperforms Gemini 2.5 Pro and matches Sonnet-4 in real-world diff edit failure rates.
I’ve been evaluating people using my @diabrowser /skill. It’s called /graham and It’s scary good. Watch me test it on a few victims live (incl. @nikitabier )
Tip: @cline is a free, best-in-class AI coding assistant for @code! It's designed for engineers who value unrestricted access to frontier AI models and comprehensive codebase understanding. On its one year anniversary, it's still my AI-coding tool of choice, with many fans at…
I was really starting to worry that I wouldn’t receive one of the agentic AI startup illuminati boxes. I bet a lot of you were worried for me too. Good news: we don’t need to worry anymore
T-11 days till we see Lee back at Vercel
I'm joining Cursor to teach the future of coding! There are millions of developers learning how to use AI and they need pragmatic advice: 1. We need to teach new developers strong foundations, so they know what to learn, and how to solve issues when debugging. 2. We need to…
Live in Cline: Qwen3-235B-A22B-2507 Open source and with a 262k context window, the latest from Qwen is impressive on the benchmarks, but we're looking forward to its real world performance. fyi, here's the model name in Cline: qwen/qwen3-235b-a22b-07-25
Bye Qwen3-235B-A22B, hello Qwen3-235B-A22B-2507! After talking with the community and thinking it through, we decided to stop using hybrid thinking mode. Instead, we’ll train Instruct and Thinking models separately so we can get the best quality possible. Today, we’re releasing…
We are experimenting with Pay-As-You-Go at Cerebras! DM me if you want first dibs at trying it
I decided to switch away from Cursor and use this deadly combination instead... try using @cline + Claude Code and thank me later! 00:00 intro 00:42 what is cline 01:06 pricing of cline 01:43 project setup 02:06 how to setup cline in vscode(@code) 02:29 how to configure any LLM…
cline discord is amazing -- we filled up our user testing sessions in 9 minutes jose's schedule is full, but if you use @cline or other AI coding tools all day everyday please shoot me a DM! would really love to chat and hear how you're using these tools. & let's grab coffee…
"all you ever do is spend time on your phone. you spend more time talking to @growing_daniel and @squirtle_says than to me." - my ex well little does she know that time was well-spent because i now have a mystery box
if you’re a twitter microcelebrity in sf you get pr packages from coding agent startups
Here's why we decided to (1) make Cline open source and (2) not make inference reselling part of our business model: When you control the inference (the AI model calls) and we build the harness (the system directing those calls), neither party can obscure what's happening. You…
it's curious how cursor and claude code are following the exact same trajectory. they start with genuine intentions to serve users, then gradually introduce "unlimited" subscription tiers and start to tilt things in their favor through rate limiting and service degradation.…
Anthropic tightens usage limits for Claude Code – without telling users | TechCrunch techcrunch.com/2025/07/17/ant…
subscription fog strikes again
there's this phenomenon happening with AI coding tools. i'm calling it "subscription fog." you have users bragging about getting $3k of inference for $200/month. then suddenly things "feel" off. when the same entity controls both: 1. the inference supply 2. the harness…