I burned a shitload of compute so you don't have to. Releasing the Harness Leaderboard.
Which harness should you run your open model in? I ran 28,849 trials to find out.
The leaderboard compares nine coding-agent harnesses on open models. In each comparison only the harness changes: the model, provider and reasoning effort stay fixed, and every task gets three attempts.
The trials ran GLM-5.3-Flash, DeepSeek-V4.1-Flash and Flint on Terminal-Bench 2.1 and DeepSWE, for $4,746 of inference.
Averaged over the three models, mini-swe-agent (@KLieret), a bash-only agent, solves 51.4% of tasks and @openclaw 27.3%. Four others are statistically tied with mini-swe-agent.
One of them is Pi (@pidotdev), at 49.4% for $0.29 per solved task, about 45% less than mini-swe-agent's $0.53.
The best harness depends on the model: mini-swe-agent on GLM-5.3-Flash, Hermes Agent (@NousResearch) on DeepSeek-V4.1-Flash and Kimi Code (@Kimi_Moonshot) on Flint.
@opencode costs $3.35 per solved task, 11.7 times as much as Pi.
See Harness Hub's Harness Lens for what the nine harnesses do differently. It documents the prompts, tools and compaction of 34 coding-agent harnesses and links every claim to its source.
Leaderboard: https://t.co/AiWiQtBOz5
Harness Lens: https://t.co/I8jbcWa9JH
Today, the calendar makes a right triangle, for the last time this century.
This Saturday is October 24, 2026.
Write it as 10/24/26 or 24/10/26, and the same beautiful relationship appears:
10² + 24² = 26²
100 + 576 = 676
But here’s what makes today special.
There are infinitely many Pythagorean triples. Only a handful fit our calendar.
Using the rule
month² + day² = (last two digits of the year)²,
the month cannot exceed 12, and the day cannot exceed 31. Those small limits leave exactly 12 qualifying dates in this century.
Today is the last.
The next will be March 4, 2105:
3² + 4² = 5²
That is a wait of more than 78 years.
An infinite family of triangles. Just twelve appearances on a century’s calendar.
Happy Pythagorean Triple Day.
A post of ours on vendor lock in and LLMs. We don't like the growing trend of AI companies quietly hiding your data while stripping away your control. We think that's bad for users and bad for the ecosystem. Here's are our thoughts: https://t.co/pM4fBX0Z1T
We've recently made Pi's cache behavior more visible. This site has been debating whether agent harnesses are helping or quietly torching their caches. That seemed like a good excuse to explain how KV caches actually work and how Pi helps (or doesn't). https://t.co/fXwUFHLAD6
used a trick @jlongster came up with
agents can control browsers but you can also ask it to record network requests into a HAR file
then it can derive a client for any website which is more efficient than browser controlling it every time
made it build a quick uber eats cli
Open weights matter less when the surrounding reasoning state remains tied to a specific harness.
Kimi K3 may be an early sign of a future where models are technically open, but practically difficult to move between tools.
Other model providers could follow the same path.
Big news: Kimi-K3 by @Kimi_Moonshot is now #1 in the Frontend Code Arena with 1679 pts, surpassing Claude Fable 5.
This is a 17-place jump from Kimi-k2.6 (#18 -> #1).
In Frontend, Kimi-K3 ranked #1 in 6 of 7 domains: Brand & Marketing, Reference-Based Design, Data & Analytics, Consumer Product, Simulations, and Content Creation Tools, landing #2 only in Gaming behind Fable 5.
The full model weights will be released by July 27.
Congrats to the @Kimi_Moonshot team on this major milestone!