Incremental gains compounds.

meta threw in the towel when it came to producing AI models since their gains couldn't keep up with China.

Has meta stopped producing new models? I figured they were just regrouping after all the drama they’ve had recently. Meta’s massive user base means they don’t need to be involved in the customer acquisition rat race. Once they have a model they’re happy with they can have a billion people interacting with it within a month.

Meta released a major new closed source model a month or so ago.

It didn't make a splash like a new open source release would have.

muse-spark is beating all the Chinese text models on lmarena leaderboard FYI. Maybe you only care about coding models.

Exactly. Go back to Opus 4.5 and see how you like it.

You won't, really.

Signaling.

Nice Fiction story.

Kimi feels almost limitless. I have the $40/month plan and I've never ever remotely close to hitting the limit. Using opus as the orchestrator.

I've had some good results with Kimi in Opencode. Can you tell me more about using Opus as orchestrator - what type of harness setup?

Skills are not like hooks. Skills can and will inevitably be ignored.

Skills are not ignored if you use a router in front of them, and they are actually called.

The problem is the base harnesses don't call them aggressively enough. Not that they don't work.

Hooks are hard stops. In theory the model must respect them, unlike Claude.md or agents.md so yeah, it helps a lot.

Yes, in theory. But these are inherently non-deterministic systems interpreting English prose. It's not the same thing as a real honest-to-God program that executes a deterministic algorithm to verify the output.

I can't believe we've sunk this low, to start complaining that the non-deterministic black box didn't respect "YOU MUST DO THIS" or "DO NOT DO THIS" commands in a Markdown file. We used to be engineers.

That has never been true.

Yes an no. Some skills are very very tuned to our own workflows. The model providers may come up with some similar alternatives but not always. Also, sometimes you need a solution now and not in three months.

I'd recommend Kimi k2.6 for your use. It is an excellent model at a fraction of the cost, and you can use Claude Code with it.

I did a 1:1 map of all my Claude Code skills, and it feels like I never left Opus.

Super happy with the results.

I was saying the same until DeepSeek v4 this morning... sorry, Kimi. The competition is intense!

Fascinated, a bummer that DeepSeek does not offer a DPA or opt-out for training. This renders it unusable for my use cases unfortunately. At least z.ai GLM has a somewhat DPA in Singapore.

The weights are open and you can use the model with any third party provider that gives you the DPA you want.

For my use-case, I want the providers to get my tokens as long as they plan to keep releasing open-weight models

If you don't use a lot of quota the cheapest monthly Claude Code is $20, Kimi Code is $19, i.e. the cost difference is minuscule.

Kimi wants my phone number on signup so a no-go for me.

What provider do you use for Kimi

The provider is a massive issue. People moving off Claude tend to assume this is solved.

Claude's uptime is terrible. The uptime of most other providers is even worse...and you get all the quantization, don't know what model you are actually getting, etc.

Kimi 2.5 was like using Sonnet 4 on a flaky ADSL line. I haven't tried K2.6 yet, but the physical unreliability of the connection was too off-putting.

OpenRouter and I'm toying around with Hermes. Seems good so far, but haven't really gotten into anything heavy yet. Though the "freedom" of not sweating the token pause and the costs not being too high is real.

Straight from them, but I know other providers like io.net can be faster but I like to directly support the project.

Thx. I'll try with my personal projects (because dues to the data collection and ToS most providers are forbidden in my company), if I can opt out of training on my input.

I'm just getting a but tired of using Opus 2.6 which eats my whole allowance and then some £££ going through the 4kB prompt to review ~13 kB text file twice - and that's on top of the sometimes utter bonkers, bad, lazy answers I'm not getting even from the local Gemma 4 E4B.

did you just copy-paste or is there a difference in the way kimi uses skills?

I don’t have the prompt at hand but basically I told Kimi (paraphrasing): I have these Claude code skills, and I know it uses different tool calls than you but read them and re-write them as your own tools.

I also created a mini framework so it can test that the skills are actually working after implementation.

Everything runs perfectly.

>I asked DS itself and it denied this

Bro, seriously?

“after evals and dogfooding” couldn’t have done this before releasing the model? We are paying $200/month to beta test the software for you.