Kinda funny that their "cost-vs-performance" chart looks the same as the one for Composer 2.5[1], except that it includes Composer 2.5 at a completely different spot.
What are the chances that CursorBench ranks Cursor's model highest, and Cognition's bench ranks Cognition's model highest? Both are to be RL'd from Kimi as a base model, BTW.
I'd posit that it's not deliberate deception, but for both companies their training data and benchmarks come from the same dataset (Devin/Cursor interaction logs) so they naturally overfit.
If you're developing on top of LLM APIs directly, this is definitely not true. There are differences in how context caching works, in what's available through native harnesses, the types of tools you're fine-tuned on (GPT uses apply_patch while Claude uses edit, with different formats), the API surface (Agents SDK, Responses API, Managed Agents), cost structures, and best-practice guidance all around.
Not to mention the meta of account limits, billing, ZDR contracts, etc.
When you're a student in a competitive program at a top university, graded on a curve, and you know your fellow classmates are cheating with AI, you have little choice but to do the same. Especially when jobs for new grads are harder to come by and there's more pressure to also go above and beyond with internships and side projects during your time in school. There's no way to compete without cheating.
The SpaceX IPO was a textbook crypto scam ICO. Launch with a huge media frenzy, super low float, "airdrop" to retail, etc.
We all know from previous token launches that the next few years will be a slow decline in price with just enough occasional bounces to keep up the hopes of the bag-holders.
You also need to consider the energy released during the big bang as a prerequisite for creating that food and gasoline. The big bang released about 10^70 J of energy, roughly equivalent to eating 10^63 big macs
According to my research, LinkedIn only does this for executive and now recruiter-like titles, but not broadly. You may be able to in order to get "verified on LinkedIn" but it's not a requirement for showing association with a company.
Yeah, all that's really saying is a weaker model with a better harness can beat a stronger model with a worse harness, specifically on the DRACO benchmark
This isn't really a surprising result. Needs more evidence to make a broader claim.
LinkedIn offers no way for $company to disavow users who claim to work for $company - they will appear on the official company page as long as it's in their profile.
We've had fake recruiters that claim to work for us running basically the same scam. These are great fake profiles: LinkedIn Premium, tons of relevant posts, etc... but they don't work for us, and we get angry messages from people saying our recruiter tried to scam them. No, they're not our recruiter despite showing up on our company page on LinkedIn. No number of reports could get them taken down.
I finally got it solved by buying drinks for a buddy of mine that works for LinkedIn, but not all startups have that connection!
What are the chances that CursorBench ranks Cursor's model highest, and Cognition's bench ranks Cognition's model highest? Both are to be RL'd from Kimi as a base model, BTW.
I'd posit that it's not deliberate deception, but for both companies their training data and benchmarks come from the same dataset (Devin/Cursor interaction logs) so they naturally overfit.
1. https://cursor.com/blog/composer-2-5