HackerTrans
トップ新着トレンドコメント過去質問紹介求人

ubermon

no profile record

投稿

Open-weight 27B hits 38% on Terminal-Bench 2.0 (Opus 4.1 hit 38% in Aug 2025)

antigma.ai
6 ポイント·投稿者 ubermon·3 か月前·9 コメント

How to Achieve #1 on Terminal Bench

twitter.com
1 ポイント·投稿者 ubermon·4 か月前·1 コメント

コメント

ubermon
·3 か月前·議論
Thanks!
ubermon
·3 か月前·議論
Totally agree! I think we are very early on discovering the full potential of local models
ubermon
·3 か月前·議論
Thank you! I think there is a lot to dive deep later with different hardware, inference engine, prompt/harness setup etc.
ubermon
·3 か月前·議論
Not yet, will conduct a more comprehensive one later
ubermon
·3 か月前·議論
[dead]
ubermon
·4 か月前·議論
and Why We Can't Have Nice Things): A Story
ubermon
·11 か月前·議論
very good articulation on the problem. My bet is that it is probably fine in the sense that the mastery itself is not going to be relevant.

I want to view AI coding as invention of new coding tools, at least in the way you described. I (hope) it will be more like punch-card->assembly-> BASIC/C -> scripting language -> some sort of well-structured natural language.
ubermon
·3 年前·議論
Give Copilot a try, it completely shift coding experience in Rust. Especially this:

> I always have to search libraries and how to do things.

Once you pass the initial curve with crutch like Copilot, then you can be almost as productive (if not more, considering refactoring and testing) with your native first coding language
ubermon
·3 年前·議論
haha was thinking the same. The other day openAI's api hiccup gave me a small panic attack
ubermon
·3 年前·議論
Moat does not come from compile time but runtime. The company, the operation, and the accumulated data, the brand, the trust and the reputation.