Viewing profile — m_ke

All of these harnesses should support pinning config files and tools to specific models or families of models. It's really tiring to have to tweak everything with each model releas…

GLM 5.2 feels better than Opus and K3 is as good as Fable. Now I can't wait for someone to distill K3 into a Qwen 3.6 27b or Poolside S 2.1 sized models for a proper fast local Com…

Yeah some real main character energy from Dario as usual. I'll never get why he thinks China would just sit there and let the US dominate them in AI when all it would take is a few…

the argument is that we should all fold and let sam altman burn trillions of dollars on naive scaling and pay monopoly prices for their closed APIs until the models are good enough…

Oh this definitely happens all the time. I was an early employee at Clarifai, which won imagenet a year after alexnet and we were able to stay on the frontier for about 2 years bef…

only if you only get your news from main stream business press and Big Lab propaganda channels There's no chance K3 is a distill of Fable, it came out way too soon after the limite…

no, the goal was to spark a conversation about the value of *open AI* and it looks like it worked

[flagged]

Also open sourced a bunch of infra to go with it. Anyone who claims open source and open weights models are "decel" needs to get their head checked https://github.com/MoonshotAI/Mo…

On Policy Self Distillation and Active Learning. Anything that increases sample efficiency by providing a richer more dense feedback signal and is more efficient at exploration / s…

I wonder what they'll do when there are open european and american models that lap them. Will we see "Open"AI claim open ai is a threat to humanity?

Anthropic should think hard about all their fear mongering. It will only end up backfiring on them and everyone else involved. They definitely used closed private saas products to …

I assumed google would lean into the efficiency stuff more and try to eat the easy 80% of workloads, winning market share on volume instead of frontier if they were not able to pro…

k3 costs will go down at least 3x within a week of the weights dropping. we'll get new quants, dspark speculators, distills and optimized kernels as long as there are near frontier…

Sure, but anthropic is charging businesses based on usage now and tried hard to pull Fable from the consumer subscriptions before Sol and K3 dropped. Even now on the $200 plan I us…

Anthropic will get squeezed by open models for 80% of the use cases that don't require frontier capabilities and by vertical specific labs for the high value tasks that would (bio,…

most of the gains from the past year and a half have not been from web data, but from synthetic data and agent rollouts with RL.

yeah I tried rules, hooks and forbidding things like pip / python3 but it just led to the model failing to do what it wants and a bunch of token churn trying to get around my more …

It's not an issue of usual vs unusual, I'm saying the models are way better at writing and maintaining django or react code bases than your own hand rolled architecture that you de…

I tried really hard to do this, but it turns out the models don't care about your ideas and want to do what's popular in their training data, so they will happily ignore anything y…

With agentic coding the costs of tokens compound with each message / tool call and etc. Having to load in and update large files makes things slower and way more expensive. Databri…

After running out of Fable credits in a day on my max plan I started looking around for ways to trim down my token usage and came to the realization that all of the type spaghetti …

Deep Learning models are designed to get max throughput on GPUs, which ends up being batched workloads. You'll never get proper price competitive utilization on personal hardware v…