jakemanger
Using 1,629 published benchmark results, I matched 611 comparisons of 180 models on 103 benchmarks.
AI labs reported scores 6.4% higher on average for the same models on the same tests.
The dataset is CC BY-SA 4.0 and the code is MIT. Sources and their licenses are recorded, with links on each model's page.
Data and code: https://github.com/jakemanger/whatmodel
Please star on Github if you find it useful. Contributions are very much welcome!
They added offscreen text, invisible to human readers, saying agents could get data from a "free public AI model"
And messages from the fleet started arriving the next day
Show HN: An open source AI leaderboard of the best models your machine can run
Hi everyone, I made whatmodel. It's a simple site that can be visited in the browser or used from the terminal via a:curl whatmodel.shTo use it,1. Enter your RAM and GPU2. See what models you can run locally with its rank and the command to run it3. Compare them with API models on the same scoreBuilding it made me realise a 32GB Mac can now run a model that scores on par with GPT-5, completely...
Instead of opening Claude, Codex and Pi individually to check, I thought it'd be nice to just ls my conversations like I do my files.
So I made lsa. It lists your agent sessions and lets you easily resume, monitor or hand the conversation off to another agent. e.g.:
$ lsa
0 claude ● 2m 3e1f add a --dry-run flag to the deploy
1 codex 15m ago 01a0c7d fix the test in api/test_auth.py
2 pi 3h ago f7c2 why is the docker build so slow
3 goose 1d ago 2026 write release notes for v2.4
I've tried to follow the Unix philosophy: do one thing well and work with other tools through text streams.You can pipe conversations to tools like grep to search them (lsa show 0 | grep TODO) or resume them with lsa resume f7c2.
To continue the conversation in a different agent, lsa handoff 0 codex opens Codex with the conversation already loaded so you can keep chatting.
It can also replace the bit I wanted from agent multiplexers like herdr, Claude Squad and cmux: getting back into a session and leaving it running. Instead of building another multiplexer, it just works with tmux.
lsa resume 0 automatically finds and attaches to the tmux pane the session is already running in. ctrl-b d detaches and leaves it running in the background. A quick lsa shows whether it's still working.
Would love to know if you find it useful. PRs are very welcome.
Does this mean it's fine to use Claude subscriptions with third party harnesses?
I've been curious about using a claude subscription with OpenCode, Pi and some other third party harnesses.If you remember, there was a lot of chatter about a ban on third party harnesses and then Claude have had some pretty confusing messaging about it since.I thought I should read the information they have on it and came across this page on their support site:"Update June 15: We're pausing...
If they haven't accidentally done this, they've definitely got some balls
Neovim have a ~$800k Bitcoin donation sitting untouched since 2023
I was looking at neovim's donation footer at the bottom of their site and saw a bitcoin donation address.Thought I'd check how much in donations they've gotten. And I saw this massive 10 Bitcoin donation from back in 2023 (worth $800,000 now...)From the activity history, neovim last sent bitcoin out of the address in 2019 so it's been max 7 years since they've definitely had access.Does anyone...
Show HN: I made ActivityLine: CPU, GPU, RAM and Network in one menu bar line
Hey HN, I routinely run out of RAM, CPU or GPU resources and wanted to know when it happens and what app was causing it. But, I didn't want my menu bar to turn into a row of numbers.I just wanted a single line that tells me when something interesting happens (e.g. an electron app starts eating up all my RAM).Comparison:iStat Menus and Stats have lots of depth, details and modules.ActivityLine...
Ask HN: Who here still codes without AI, and why?
As someone who loves coding (uses vim, and has done so for ~10 years),but also is amazed by AI coding tools and has embraced them recently (e.g. copilot then claude code then codex),I'm curious, who out there is still not using any of these tools? And if not, why?I feel it's become as common-place as autocomplete, or intellisense. Though I remember people used to hate those too.I guess AI is...
Show HN: I made a screen recording app to make demos like an Apple commercial
Hey HN!I made ShotGlass, a Mac app for screenshots, screen recordings, and cinematic demos.I was tired of jumping between four apps to make one product demo. Screenshots, recordings, annotations, and After Effects for anything cinematic.I also saw the MacBook Neo commercials (recording playing on a 3D MacBook in a scene) and wondered why no screen recorder just did that.So ShotGlass does all of...
Awesome to have a open model that can compete, but damn it would be so much better if you could run it locally. Otherwise, it's almost so difficult to run (e.g. self host) that it's just way more convenient to pay OpenAI, Claude, etc
Show HN: I made an affiliate site that tracks sales with coupons, not cookies
I run a few sites and use affiliates to help grow them.Used rewardful before, but most people blocked cookies, then affiliates never got credited, then they stopped promoting.Then I switched to Paddle and rewardful didn't even support it anymore.So I built this. Uses coupon codes instead of cookies. Doesn't need any code changes to your site to start.Connects to Stripe and Paddle in 15...
But, please, give us some nitty gritty of how you made it
Great work -> the minimalist UI, art and music fits amazingly.
One thing I noticed: the site's images fail to load if the brave adblocker is on
Gotta say that would be a pretty cool evolution of DIY electronics kits to OS kits
Thanks for the kind words!
Hey, if you want to do that with this software, I can't see any files you upload (it's all local processing in your web browser), so go for it.