dools

Yeah you get the usage back and each element has pricing published

It’s not meaningless at all: every query returns usage and I can calculate the cost.

EDIT: this is like saying hourly rate or salary is meaningless. Different people have different output. You have to evaluate performance.

EDIT2: just pray the LLM providers don’t start taking Patrick McKenzie’s advice and start charging based on “value delivered”

Orly?

“ Automatic Prefix Caching (APC in short) caches the KV cache of existing queries, so that a new query can directly reuse the KV cache if it shares the same prefix with one of the existing queries, allowing the new query to skip the computation of the shared part.”

https://docs.vllm.ai/en/latest/features/automatic_prefix_cac...

I would be very surprised if they hadn’t sorted out some form of shared KV caching

I tried a pro model out the other day and thought there must have been a bug in Pi’s cost calculations. But no, it’s absolutely fucking insane. Wasn’t even any better at the task.

No, PDO doesn't qualify, because it lacks the "relational mapping" part. If I want an AI opinion of my project I can always ask a chatbot myself.

I dunno... Command and Conquer ported to macOS/iPad in 18 months using AI coding agents probably wouldn't have the same ring to it

No, it's an ORM because it gives you object based iteration over your query (and the ability to use custom classes for those objects, you just don't have to create classes for every single thing if you don't need them).

EDIT: oh wait looks like I never got around to implementing the ability to use custom classes :) this is still in the to do section:

come up with a good "mix in" style to cast the objects returned from the iterator to a new class for implementing custom functionality (that one would normally include as part of the "boilerplate" class)

Yeah which is a bit underhanded. Because the implication of "using Fable" is that it was done in under ... a week? So it's just a bit of click bait.

How is it done "using Fable" when the first commit was Feb last year??

I always disliked ActiveRecord, but I figured ORMs don't have to be ActiveRecord. I created this library 14(!) years ago not too long before this article was written https://github.com/iaindooley/PluSQL

The idea is that you like SQL, but it gets repetitive writing joins and accessor code. I had always hoped it would catch on as a pattern: no boilerplate, automatic mapping to objects in your code of any query (whether generated by the ORM or passed in as a raw query) and easy to override/dynamically build bits of the query as you pass the object around.

> When you don’t have an established wired infrastructure, it becomes significantly easier to jump technology generations

> jump technology generations

Satellite internet is not a “generation above” fibre internet

Yeah I’ve been consistently underwhelmed by anthropic models, but then I don’t use their harness so maybe that’s it

Why would a Chinese firm distilling the product use Claude code?

I think Opus 4.8 is deliberately nobbled. Kimi k2.6 with Kimi code beats opus models at finding vulnerabilities, even though it produces some false positives, when I give the same issues to opus and ask it to verify most of the time it concurs it’s a real issue even though it failed to find the issue itself

I've been building a reasonably complicated project over the past week using deepseek v4 pro almost exclusively (a couple of k2.7 and 1 session with gpt5.5 to re-assess some architectural questions). Deepseek is super capable though if you're a coder. I don't even write "code" but I can tell when it's doing something dumb and tell it how to do it better, but other than that I'm not micro managing it or using it "just for auto complete" or whatever.

And it is SO fucking cheap.

Bro just give me a new iPhone mini

> You aren't required to identify yourself to get a phone. You can get a prepaid phone with no ID.

Not in Australia

How is it any different from being required to identify yourself to get a phone or electricity account? Identifying yourself on the internet is long overdue.

Yeah I mean like git clone the repo then "hey LLM rip off this code, make no mistakes"

Vibe stole it?

Kimi k2.6/7 running inside Kimi code already kicks the pants of the latest Claude and OpenAI models when it comes to cyber security. I regularly run multi model security reviews and while opus 4.6/7/8 and gpt 5.3/4/5 find a couple of things and declare mission accomplished (running inside pi) kimi k2.6/7 inside pi finds more issues and inside kimi code finds the most.

There are sometimes false positives but when I give Kimi’s report to the frontier models they more often than not confirm they are valid security issues but didn’t find them themselves.

Compared to what comes out at the end. Like if you sit there watching Kimi k2.6 "think", you're like "what? no you fucking idiot!" and you get this urge to "steer" it and so on, but very rarely is that steering actually necessary, it just winds up popping out the correct answer and all of those 'Wait! That's it! I found it! Actually ... Let me just' is just whatever internal processing it needed to use to get to the correct response. Mostly likely it's just being self-adversarial and exploring a bunch of dumb avenues to isolate the best outcome with the highest probability

The reasoning traces always look terrible and they’re frustrating to watch. It’s the same with Kimi. What’s interesting is that the end result is then good. I think it’s just some sort of devils advocate trick to get better output.

Is z.ai

Is 2 better than x.ai

The new Google ones don't look too bad.

If I need context for a session then that is output from a previous session, otherwise I find any “memory” functionality cumbersome.

I saw /graphify recently which cuts down on exploration cost and seems more appealing (although I haven’t tried it yet)

This was in my staff documentation and it’s now in my AGENTS.md: tell don’t ask.

If there is a decision that you need to make don’t ask me for input, do the thing that you think makes sense and then write down what you did and why.

If it’s the wrong thing I’ll update the docs to make it clear for next time.

Without this I would always wake up in the morning to an inbox full of questions and no work done, rather than an inbox full of finished tasks and maybe a couple of corrections.

With LLMs if I ask for a code analysis and plan to fix something they tend to put a list of questions at the end about which they want confirmation.

Then I have to waste time saying yes or no or coming up with the solution. If I tell them to instead just make assumptions and record them all at the end then I only need to correct 1 or 2 assumptions if required.

The idea of giving a non deterministic automated process direct deployment control is fucking madness to me. That’s why I don’t get the obsession with MCP. Deployment can be scripted. It doesn’t need an LLM, it is a completely deterministic process and you want it to run identically every single time.

The right model for agentic API usage is having LLMs write scripts that use APIs. Connecting agents to MCPs and telling them to go and do stuff over and over not only wastes money but invites catastrophe.