Lachlan

One of my favourite ways to get a vibe for AI models coding ability (particularly it's ui preference) is via @lmarena_ai they have their WebDev arena which pits two AI's against each other, blindly and you pick which result you like the most. You begin to learn the styles that particular AI uses, it helps you understand why people generally prefer Anthropic outputs. Both their 3.5 models, Sonnet & Haiku share a similar design style which is more aesthetically pleasing for sure. I was also impressed with some of the new Gemini 2.0 Pro outputs, and Qwen Max, these are definitely a leap up compared to previous gen models also. But still just behind sonnet 3.5 overall, generally* I was generally underwhelmed by the O3 mini high outputs, (particularly UI). *It's a very useful exercise to show you that there isn't one "best" model, even though it's easy to say "Claude is the best" - Each prompt is different and getting opinions from other models is very useful.

I quite like the model its performing quite well in our harness compared with the preview model, however the 3x in price is hard to justify / changes the utility of the model significantly. if the pricing hadn't been so much higher i think the reception from people would have been a lot better.

Gemini 3.5 feels like the start of a new era for Gemini, we spent the last 2.5 years putting the infrastructure, products, team, etc in place (learning lots of lessons along the way). The model is the product, please keep the feedback coming!

@zachtratar Yeah pretty crazy how many security issues we are seeing. Was just thinking the same thing - x.com/lowcodelocky/s…

With the amount of supply chain attacks across different packages developing with Xanoscript keeps making more and more sense.

@swyx /goal keep working until I say stop..

@samhogan Nice work Sam, having a play now!

We’re introducing HALO 😇 Hierarchal Agent Loop Optimizer HALO is an RLM-based agent optimization technique capable of recursively self-improving agents by analyzing their execution traces and suggesting changes. This work is inspired by the Mismanaged Genius Hypothesis proposed by @a1zhang and @lateinteraction earlier this month. tldr; we improved performance on AppWorld (Sonnet 4.6) from 73.7 --> 89.5 (+15.8) by giving HALO-RLM access to harness trace data and asking it to identify issues. The feedback from HALO surfaced failures in the harness such as hallucinated tool calls, redundant arguments in tools, refusal loops, and semantic correctness issues. Each issue mapped cleanly to a direct prompt update. We then fed these finding into Cursor (Opus 4.6), and asked the coding agent to update the underlying harness. We repeated this trace -> HALO-RLM analysis -> code update loop until the score plateaued. Today we’re open-sourcing the core HALO-RLM framework, evals, and data for further review.

@matvelloso Agree it's a painful process - surprisingly easy for google though via their @GoogleAIStudio they integrated their AI auth / db services quite well.

I invite you to try to create a simple, functioning OAUTH app on both Google's and Microsoft's identity systems. Just that. Start with nothing, and count how long it takes.

@ivanfioravanti @0xSero yah sucks at coding unfortunately

@0xSero But in coding... better to avoid it as much as possible.

This is god tier and fits on 24gb + It crushes everything up to 6x its size and ties with Gemini deep think and Deepseek on Math huggingface.co/nvidia/Nemotro…

Going to be a fun week of launches : )

These benchmarks look a lil too good to be true, interested to hear how it performs under real-world usage.

@joshwoodward @OfficialLoganK Why are you only able to revert / edit your most recent prompt / message in the gemini app? You also aren't able to roll the conversation back to a previous point. Seems like an easy win to give users more control. Like in AI Studio.

New @opencode plugin incoming. Your safety is our mission.

@ptsi @claudeai Yeah only appears to work in the terminal - would be awesome to get support for it in the extension too!

@claudeai It seems like Claude-in-Chrome MCP is on by default in CLI but not VS Extension. So it still isn’t accessible or has to be configured manually somehow? Unless I’m missing something? 🙏

The VS Code extension for Claude Code is now generally available. It’s now much closer to the CLI experience: @-mention files for context, use familiar slash commands (/model, /mcp, /context), and more. Download it here: marketplace.visualstudio.com/items?itemName…

@codewithantonio @opencode I moved when claude-code had a bug with their compact feature and made the model stupid - haven't gone back as opencode is doing the trick.

for those of you who moved from claude code to @opencode , what was the main reason?

@OpenAI Please update your docs and in-app experience regarding building apps with a free vs paid account. There's no mention a paid account is required, developer mode can be enabled on a free account and building one in app presents no error, it simply just doesn't create a new app. Please throw an error letting the user to know they need to upgrade and mention it in the docs - wasted enough time!

use code "FEEDBACK" for 3 months free on greptile valid until sunday DMs open

@MoonDevOnYT The bait is strong with this one

if you are spending more than $20 per month on claude code, openai or any ai you literally hate money there is this service called ollama it allows you to run ai on your computer buy one of these acer laptops for $210 (probably can get it for under $200 for holidays) download ollama and install your favorite ai free ai for life you're welcome bonus - use a credit card and get the payment on the laptop to under $20/mo

@MoonDevOnYT Mac Studio not mini

@MoonDevOnYT More likely you’d be wanting two of these at least (this is AUD not USD) but you’d be well above 1K USD a month to start handling the bigger models with larger contexts

Before you pay $200/month for another month of Claude code... What about just allocating that $200/month to Apple with a 0% APR? Running a local model in a local agentic coding thing like CC Ironic that the most souped-up m4 Mac Mini comes out to almost exactly $200/month, which is the same price as the Unlimited (Limited) Claude Code. The only questions pending are, what model would you run, and what Agentic coding thing comes close to Claude Code?

@opencode I really hate the way you have conversations setup in the mac app - such a frustrating experience using it tbh.

@scaling01 I'm enjoying it at the moment, i've got more confidence with opus 4.5, but so far for micro adjustments / refining things it's been very solid. The quick iterations keeps you in the flow.

Gemini 3 Flash scores higher than GPT-5.2, Opus 4.5 and Gemini 3 Pro on SWE-Bench Verified ???