El Combo

cursor is the kind of company where you complain about something to someone and they’re like oh i built a prototype for this thing that solves that exact problem, let me add you to the flag today i was on BOTH sides of this happy to be here

@kunchenguid The Cursor approach IMO is more about how to intentionally configure their auto mode with a balance of cost vs thinking type controls. I don't see it doing much more than that at this stage. I like your take! So fat my testing is the intelligence option is chewing tokens.

maybe this is controversial, but i believe what Cursor shipped here is a wrong solution to routing intelligence more generally, any attempt to do model routing at request level, while may yield some small gains, is fundamentally flawed and doomed to fail here's why - the complexity of a task only reveals itself when you start working on it. this is the same reason why we humans are often wrong when asked to give cost estimates upfront the correct solution is have a smart model (often a tech lead in human teams) do some planning and understanding, and hand over the implementation to another agent with appropriate level of intelligence and reasoning effort when the task is delegated to a less intelligent model, the smarter model also needs to continuously monitor the execution and examine outcomes to ensure things are on the right track this is a system that proved to work really well in firstmate and helped me save a lot of tokens. routing should work at the boundary between agents and subagents, not per each LLM request

Introducing Cursor Router, our intelligent model router that selects the right model for the task at hand. Router delivers frontier-quality results at 60% lower cost.

@databricks I wish this got ported to the largely abandoned Redash project...

Building beautiful, on-brand dashboards just got easier 📊 Dashboard and workspace themes in AI/BI let you customize fonts, colors, and visualization palettes for a cohesive experience. Apply a theme to one dashboard, or set it once at the workspace level and have it carry across every dashboard. See how it works: databricks.com/blog/design-be…

Gemini 3.6 Flash is now available in Cursor!

🚀🚀

introducing OpenBench v1, an open framework for measuring AI performance and efficiency for your codebase and use case. Companies are realizing that you can't simply tokenmaxx, and are frantically looking for better ways to use and measure AI use and efficiency. One approach is evidenced by the rise in model routers: OpenRouter, Cloudflare, Databricks, Vercel, and now Ramp to name a few. But they'll also need the ability to evaluate how agents are performing in their actual use cases and codebase. A good example is @DoorDash's recent evals on their codebase and prs. This will become even more important as companies explore different model + harness combinations. OpenBench will be built to this direction, focusing on the cross between correctness vs efficiency in both token use and latency. Starting off, OpenBench makes it easy for anyone to add to the task set, and run a variety of harness + models. The framework also makes it easy to add any custom harness variant, for example I've been testing codex variants with ablations against the stock codex harness. OpenBench will also have tooling to help with discovering/adding reliable tasks from your repo. Today I have integrated harnesses: codex, claude, cursor, devin, grok build, pi, and eval'ed them for different models like gpt 5.6-sol, GLM 5.2, Kimi K3, Grok 4.5, etc. measuring correctness %, token use (in/out/cache), and latency.

@tibor_tee Not sure what's more impressive, that this is possible or that the docs were good enough to be used.

Incredible agent swarm research

First day after the World Cup and not having to wake up early to watch the games. Only to have to wake up early for a flight. Going to sleep all weekend. Who knew watching sports was so exhausting.

Qwen3.8 is launching and going open-weight soon!🌐 With a massive 2.4T parameters, this model is continuously evolving. We believe it’s one of the most powerful model available today, compatible to leading frontier AI models , second only to Fable 5. You don't have to wait to test it. Just now, the Qwen3.8-Max-Preview made its debut on Alibaba’s Token Plan, Qoder, and QoderWork. Be among the very first to try it out. Can't wait to hear what you build. Stay tuned! 🚀  Token Plan international：qwencloud.com/pricing/token-… China：platform.qianwenai.com/pricing/token-…

5.6 Terra high is underrated. Switched @clawsweeper (GitHub review bot) to it and it's ~40% faster overall with negligible quality loss. Better than 5.5 on all counts. Massively cheaper. (Tried xhigh but that negates perf wins, didn't make a noticable difference in review evals)

Quick, everyone act surprised! This was bound to happen. You can thank Grok 4.5, GPT 5.6, and Kimi K3 for this.

New flow I've been enjoying 1. "/grill-with-docs <issue description>" 2. "Oh damn, this is way bigger than I expected" 3. "/wayfinder make a map of this" 4. Continue happily on

@opencode What about Zen users?

2x the kimi k3 usage on opencode go for 1 week

@mattpocockuk We should do that with people and work too.

Please, please, please when you tell someone the model you're using also say the effort. Every SOTA model acts completely different per effort level.

I don't like to fanboy over large companies. We've used Claude Tag for over a week. It's the first time I've experienced AI in a collaborative context with the team. The integration in Slack makes it a true collaboration partner. @claudeai @AnthropicAI oustanding 👏

@leerob When in cursor? @leerob 🙄

New open model from Thinking Machines! 1T MoE, 1M context, multimodal, some solid evals. Really nice blog post 👏

@birch_js @OxcProject It's been great for us so far. Only gap is lack of oxlint support for graphql. I could use the alpha js eslint plugin but don't want to run that in production until it hardens. It's the only gap making me consider biome at least for gql to get off eslint.

Just migrated from eslint + prettier to oxfmt + oxlint. - Linting 724 files: 26s -> 2.3s - Formatting 1,028 files: 14 s -> 3.8s That's a speedup of 11.3× for linting, 3.7× for formatting! 🏎️☁️☁️☁️ Bravo @OxcProject 👏

We just doubled the included usage of Cursor models on all plans. Enjoy more access to Grok 4.5 and Composer 2.5!