Astra writes well (which is why I often like to work with it over Claude), but it tends to produce code golf that's borderline unreadable, like completely omitting line breaks. I thought I had something funky with my config, but it seems to be a known issue.
The generated code certainly does what's asked of it, but it's as sloppy as anything else I've seen. I don't think this benchmark can control for that. Maybe if it had a metric for minimum diff to satisfy the new requirement with the existing constraints on dead code etc?
I have a simplify skill built from a rich hickey talk that basically solves the erosion and verbosity metrics SCB flagged (i.e. erosion and verbosity are way better and improve over time). However, after running the SCB benchmark I cannot find any evidence that it actually improves test pass rate trajectories. Food for thought on maintainability metrics.
Is there a reasonable way to run a representative subset myself?
Edit:
> For folks who want to try to repro these findings: I used the pinned v1.0 catalog, the stock just-solve prompt, seed 42, the any-case runner pass policy, and Docker with Python 3.12 and uv.
This is interesting. Is nobody training LLMs to repeat resolving issues by patches from real projects where refactoring is required, judging the refactor similarity and not just tests passing?
For most things I find medium to be the highest quality output. However if I want it to do spatial reasoning or if I want it be funny / write something (not code) of almost-passable quality, then xhigh is where it's at.
I believe effort can mean both "solve harder problems" and "produce more stuff". I mostly don't want more stuff than I asked for, so a capable model at lower effort can indeed be the wiser choice. Faster response too.
Secondly, I’m mostly using Astra Ultra or Sol 6.1 when tight on quota. I pay $200/month and rarely if ever run out. Am I doing something wrong? Or perhaps I’m not that productive as you guys?
Same here. I’m feeling productive enough. If I was more “productive”, I wouldn’t be able to keep up with what I was producing. My attention and cognitive broadband is the bottleneck. I find it hard to believe that people are actually productive using 4 active chat windows simultaneously for 8+ hours a day.
Astra writes well (which is why I often like to work with it over Claude), but it tends to produce code golf that's borderline unreadable, like completely omitting line breaks. I thought I had something funky with my config, but it seems to be a known issue.
https://lucumr.pocoo.org/2026/9/7/astra-why/
[Sorry for X] https://x.com/tenobrus/status/2096636223414808719
The generated code certainly does what's asked of it, but it's as sloppy as anything else I've seen. I don't think this benchmark can control for that. Maybe if it had a metric for minimum diff to satisfy the new requirement with the existing constraints on dead code etc?
Don’t apologize for x, just don’t post x links.
Wouldn't it be better to enforce pass via the harness and then measure cost/quality since that's more what people are likely to do?
I have a simplify skill built from a rich hickey talk that basically solves the erosion and verbosity metrics SCB flagged (i.e. erosion and verbosity are way better and improve over time). However, after running the SCB benchmark I cannot find any evidence that it actually improves test pass rate trajectories. Food for thought on maintainability metrics.
Is there a reasonable way to run a representative subset myself?
Edit:
> For folks who want to try to repro these findings: I used the pinned v1.0 catalog, the stock just-solve prompt, seed 42, the any-case runner pass policy, and Docker with Python 3.12 and uv.
And https://github.com/gabeorlanski/scb-problems/
This is interesting. Is nobody training LLMs to repeat resolving issues by patches from real projects where refactoring is required, judging the refactor similarity and not just tests passing?
For most things I find medium to be the highest quality output. However if I want it to do spatial reasoning or if I want it be funny / write something (not code) of almost-passable quality, then xhigh is where it's at.
Are the raw outputs published somewhere?
I am currently investigating SCB results and a larger, more representative dataset to work on would be really helpful!
Great benchmark!
I believe effort can mean both "solve harder problems" and "produce more stuff". I mostly don't want more stuff than I asked for, so a capable model at lower effort can indeed be the wiser choice. Faster response too.
First of all, what happened to Max and Ultra?
Secondly, I’m mostly using Astra Ultra or Sol 6.1 when tight on quota. I pay $200/month and rarely if ever run out. Am I doing something wrong? Or perhaps I’m not that productive as you guys?
Same here. I’m feeling productive enough. If I was more “productive”, I wouldn’t be able to keep up with what I was producing. My attention and cognitive broadband is the bottleneck. I find it hard to believe that people are actually productive using 4 active chat windows simultaneously for 8+ hours a day.
Related. Others?
Benchmarking Opus 5 on SlopCodeBench - https://news.ycombinator.com/item?id=49076391 - July 2026 (119 comments)
Finally a serious benchmark!
I wish a company named human layer won't write their reports with so much Claudisms.
You expect an AI company to slow down and care about what they produce? How naive are you?
If you haven't pushed so much that people are physically unable to look at it, are you even running a company?
I think people downvoting you missed the sarcasm.
On the contrary.
The irony is the article is very sloppy. Can we at least write it and review it properly..
What's wrong with it? Reads fine to me.
This is true. I was unfamiliar with this concept (I use codex/claude daily), and am trying to get to the "bottom line".
Oh, it's there at bottom under "my takeaway". But I still didn't get it initially.
Slop Code Bench? Need a Slop Writing Bench too