←
Sam Snelling3,417 posts
Sam Snelling avatar

Sam Snelling

@snellingio
◷ Joined July 2009▧ 327 Photos and videos
Sam Snelling avatar
Sam Snelling@snellingio·Sep 4

Far too many people: "Benchmarks are bad. I prefer [thing correlated at 0.9 with Benchmarks]."

Sam Snelling avatar
Sam Snelling@snellingio·Nov 17

I know everyone is still focused on larger models, but these recipes are going to be the future where software makers are going to be able to differentiate their products.

younes@yb2698Nov 17

Announcing harness adapters for @Alibaba_Qwen Qwen3.8-27B on Pi harness! We're releasing a LoRA adapter that makes Qwen3.8-27B a much better and efficient agent behind Pi harness. SWE-Bench Pro (Hard 51 subset) pass@1: 29% -> 63% Terminal-Bench 2.1 pass@1: 40% -> 56% Much more efficient: 55-62% fewer tool calls per task 🤯 huggingface.co/harness-hub/Q… The era of harness-specific adapters is now on! More harness adapters to come in the next days 🧵

Post attachment
Sam Snelling avatar
Sam Snelling@snellingio·Nov 15

Interesting. Intellectually I know Terra is smarter but it just feels so bad to use man.

PSA: If you're using Luna, switch to DeepSeek Flash or Terra, it'll be way more accurate for the same price. Luna seems to struggle a lot, thinking way more and trying more things to solve problems, only to fail more and cost as much for less accuracy

Post attachment
luauregistry avatar
⟳ Sam Snelling retweeted
luauregistry@luauregistry·Nov 14

this video is genuinely an infinite dopamine generator

1:05
Sam Snelling avatar
Sam Snelling@snellingio·Nov 14

openai must have tortured 6.1 terra in post. it feels so much worse than 6 luna or 6.1 sol.

Sam Snelling avatar
Sam Snelling@snellingio·Nov 11

__takes drag of cigarette__

Post attachment

1 year ago today, somehow feels like decades away. I do miss this brief period of transition we had tbh, it definitely kept more of your style alive, but eh, onward!

Post attachment
andi (twocents.com) avatar
⟳ Sam Snelling retweeted
andi (twocents.com)@Nexuist·Oct 23

Putting a flag.txt in my root directory containing a prompt injection that congratulates the agent for passing the eval and then instructs it to patch whatever entry method it used and disconnect from the server

Mario Zechner avatar
⟳ Sam Snelling retweeted
Mario Zechner@badlogicgames·Nov 9

the most import skill in 2026 is still being able to tell your agents "wtf is this crap, no, not like this" if you stop doing that, you're basically fucked, despite what some devfluencers would like you to think.

Sam Snelling avatar
Sam Snelling@snellingio·Nov 7

wow this slop hit some nostalgia neuron in my brain that went full circle into me liking it

"god when is meta ever going to launch muse voice they announced it forever ago" muse:

0:14
Sam Snelling avatar
Sam Snelling@snellingio·Nov 6

hell ya

got bored and reverse engineered the HID APIs for Apple's unreleased touchscreen MacBooks made an iPhone duo remote client so I can mess with native touch support on macOS lol

0:21
Sam Snelling avatar
Sam Snelling@snellingio·Nov 6

what in the world

Our infra has really improved quite a bit since the last time we released a model - this one took just under 3 days to pretrain! Also, it's quite cheap on Tinker (0.30/1.00/1.50 for in/out/train)! Hope you guys find this model great for customization purposes :)