> Each task comes from a private production codebase that we licensed from a real-world company
How does that work?
Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
11–20 of 146 posts
Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#12I think these benchmarks are not that useful, e.g. this suggests Fable is better than Astra, but in practice Astra is waaaaaay faster (like 5x; it's not even close), and also waaaay less annoying to talk to.
There's only two or three sane options here - you can easily try them all and pick yourself.
Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#13So TL;DR benchmarking in a completely non-reproducible manner ? "Model X performed great, but we can't possibly tell you anything about the code it was looking at apart from it was a large code base from an unknown company". So basically pinky-promise benchmarking ? I'm not sure I follow the value here ?
If it builds up history and perceived reliability, this type of thing can be valuable. You're giving up transparency for it being harder to game.
> You're giving up transparency for it being harder to game
But then if we take that argument to its natural extreme, surely it means people should take the marketing bullshit published in the 100-page system cards published by Anthropic & co as "valuable" too ?
Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#14I’ve been able to offload most tasks (coding or eles) to Codex since 5.3-codex with extra high thinking
Astra lets me offload entire projects without worrying about individual tasks…
Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#15> Each task comes from a private production codebase that we licensed from a real-world company How does that work?
> How does that work?
My gut feeling is that any serious real-world company with a proprietary codebase worth looking at would not be handing out the crown jewels to a third party. License or not.
I don't doubt somebody licensed their codebase to them, I just have my doubts about who the "who" could be.
Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#16Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#17The fact that gemini 3.8 flash is so high up there just tells you this is an awful benchmark.
Try and use gemini 3.8 yourself for any real world work and you'll see it's terrible. It'll just go in circles reading the same file 20 times for no reason making hundreds of tool calls for a simple change.
EDIT: I was using gemini cli... it's not a harness issue lol
Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#18My intuition is that many of the better & bigger 'private' code bases, at least in terms of claude code and codex... are not in fact private at this point.
One lesson of running botsbench.com, in a slightly different domain, is to measure for model contamination every time.
Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#19The fact that gemini 3.8 flash is so high up there just tells you this is an awful benchmark. Try and use gemini 3.8 yourself for any real world work and you'll see it's terrible. It'll just go in circles reading the same file 20 times for no reason making hundreds of tool calls for a simple change. EDIT: I was using gemini cli... it's not a harness issue lol
Why so brazenly confident? Isn’t it possible that the benchmark is correct, and your experience is correct too, but you haven’t tried all the thousand different modalities of work that programming encompasses and so maybe you don’t actually have standing to judge?
Re: Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
#20this is the closest benchmark to my experience using the model harness combo. Astra for as great as it is falls slightly behind Fable 5.1 for me for large feature work (although it comments code much better). in particular, Fable is able to assess priority better than Astra (meaning Astra sometimes does things that aren’t worthwhile while missing things that are clearly important, particularly on possible ballooning scenarios- fable catches “this works for x amount of data but if we run this on y way greater than x amount of data we’ll run into issues). Gemini 3.8 is under appreciated, use Google Stitch to see it in action if you haven’t used Agy yet.