sieve

MS13 is pretty sharp and has been my workhorse for the past month. It follows my coding style and commit/clean workflows referenced in AGENTS.md perfectly but has the habit of doing things without conferring with me (the Gemini problem). So you need some kind of instruction for that.

It starts failing around the 5-600K context mark, but you can have it generate a handover document and continue in the next session.

I would not use it at sticker price, but the Contributor version is priced just about right.

I am not designing rockets. Most of my work is bog standard hobbyist stuff: compilers, vms, sandboxes, system tools of various kinds, SSGs, markup languages, plain text ledgers etc. Even Gemma/Qwen running locally can manage this.

Frankly, I have no idea what people do with Opus/Fable etc. I don't think anything I do needs something that charges $50/M for output tokens.

Are they bleeding? Their multipliers seem to be reasonable. They are not offering $60 worth of usage for $10 on every model, only some. In the case of the expensive ones, it is only $15.

Given how subscription models work (not every one uses every last $ of their plan), they should achieve breakeven soon enough I guess.

My OpenCode Go stats for the last 30d:

Cached Read: ~6,500M

Input: ~150M

Output: ~20M

Approx $40 worth of usage across DeepSeek V4 Flash + MuseSpark Contributor 1.3. And a bit of both the GLM models. This is covered in a $10 subscription.

If I were to use Luna's API pricing:

$0.02 x 6,500 = $130

$0.20 x 150 = $30

$1.20 x 20 = $24

So $184. And this is assuming smaller coding sessions (<272K) beyond which Luna pricing doubles.

--

Cost wise, these models are nice for small stuff. Translations etc. Any model that does not provide multiple Mtoks of cached reads per cent is not very useful to me for coding workflows.

I guess people are slowly realizing that giving LLMs r/w access to your entire machine is an utterly insane idea.

I did a quick look-around last month and decided to start using bwrap. But manually configuring it on a per-project basis is irritating. So I rolled out something for my own use (+ a couple of friends) based on bwrap.

What I do:

- start with `bwrap --clearenv --unshare-all --die-with-parent --tmpfs / ...`

- every single file and folder and envar I need has to be mapped in. I have profiles in TOML, and `prepare/probe` commands to make this task simpler

- `--tmpfs /` means sandbox inits as `/home/user` on tmpfs unless you specify your own `home` and `user` keys.

- `--unshare-all` means there is no network inside the sandbox. So I used `socat` to run a HTTP/S proxy inside. Lets me control exactly which host+port combinations can be accessed. But this means nothing except HTTP(S) works. So no ICMP/UDP/TCP.

- profiles can be extended via extend syntax (otherwise you have to prepare/probe/manually specify everything per profile which is a nightmare). Which lets me do a base -> net -> coding chain.

Code is merely the means to an end. It does not matter who writes it as long as it does what you want. Correctly.

Stands to reason though that someone who knows programming AND has domain knowledge can get LLMs to produce much better output compared to someone who does not.

I never used to have time to make all the stuff I needed or was interested in. With LLMs, I can.

This "shipped a whole app while being in the gym" does not work for me though. It takes me a couple of days to a week to produce solid, functional software (~10KLOC). Simple tools (3-400LOC)? Yeah, those you can produce in 30-60 minutes.

I became interested in sandboxing last month after watching LLMs fail to respect basic boundaries. Well, the very expectation that they would is foolish in the first place.

I am not a fan of application-level sandboxing. The JVM tried with its security manager, and Deno with its allow/deny, but it is not general enough for me. At some point you have to assume that anything you run on your machine is possibly broken/compromised and then deal with the situation depending on your risk appetite.

This is a long story that I have written about on my blog, but I decided to go down the Bubblewrap + seccomp + socat route for the sandboxing tool I built. Let's me run harnesses and compilers and even headless Firefox in sandboxes without worrying about damage to random parts of my system.

Yes, I meant the Flash version.

I have used Kimi 2.5 and GLM 5.3 (& 5.3 Flash). Do not need them for what I do outside of spec hardening (basically, a lot of chatting).

I tend to know exactly what I want and most of the weaker models are enough to get me there. I have mainly been using MiMo, DeepSeek V4 Flash and MuseSpark Contributor over the last month or so.

I regularly hit 200-300M cached reads every day on some of the models I use. It has exceeded 7-800M on a couple of occasions. At $0.04/M, that is $8-12 per day only for cached reads.

> Basically the old theory is true - the code IS the specification.

The spec is whatever I write by hand. The code is what the LLM writes for me. The spec could be anything depending on how much detail you want.

The problem with the "code IS the spec" in the age of LLMs is that they will change stuff without telling you while hitting their immediate goal. Six months ago, I used to review every single change. Now I get the LLM to audit the code to compare against the spec. Any divergence means one of two things:

- either I have to update the spec, or

- the LLM has to update the code.

Lots of things!

- Wanted to start blogging. So built an SSG for that.

- Have an ink tank printer that must be used a few times each month or bad things happen. Have been printing Sanskrit stories instead of test pages. Was manually building booklets with typst and then using pdfimpose. Finally, decided to write an app that does MD -> booklet.

- Decided that running harnesses directly on my machine is a bad idea. Docker/Podman are too complicated for the task at hand. So I built Adamant. It started out as a basic wrapper around Bubblewrap. Then I added networking via a socat proxy. Now it can do this:[1]

  # Runs a webserver inside the sandbox that is accessible from outside
  adamant --profile minimal prepare -- ls sh fish ncdu fastfetch uname mkdir python echo
  adamant --profile minimal probe -- python 'print("Hello, World!")'
  adamant --profile minimal run --ingress 45678:45678 -- fish -c "mkdir -p /tmp/www; echo '<p>Hello, World!</p>' >> /tmp/www/  index.html; python -m http.server -b 127.0.0.1 -d /tmp/www 45678"

  # Runs headless Firefox inside the sandbox and takes a screenshot
  adamant --profile ff prepare -- ls sh uname mkdir fish firefox
  adamant --profile ff probe --timeout 60 --host bhashika.org.in -- curl https://bhashika.org.in
  adamant --profile ff probe --timeout 60 --host bhashika.org.in -- firefox --headless --no-remote --screenshot https://bhashika.org.in
  # screenshot.png is produced in $PWD
  adamant --profile ff run --host bhashika.org.in -- firefox --headless --no-remote --screenshot https://bhashika.org.in

- Building an actual harness called PonderCode for personal use that uses ideas similar to Adamant as I find TUIs irritating for the text heavy work I do. Copy-pasting is a nightmare as almost everything is space-padded.

- Building a language learning product for Indic languages. Monetization is difficult due to the tiny market plus self-imposed restrictions like "only serious learners/readers need to pay." Only time will tell if it works out.

- My Python-replacement VM+PL project is on a bit of a hiatus. Will probably revisit it in a couple of months. Had made a lot of progress in April till I decided to expand the scope and got burnt out. Never do that.

[1] https://bhashika.org.in/logs/2026/adamant-devlog-3

You have to wonder why someone did not think of this form factor sooner. I have the original X4 and have been enjoying it a lot. My Kobo mostly lies unused because I do most of my reading in bed and the X4, even with a clip-on light, is so much easier on the wrists.

Dual core Arm CPUs, Android etc used by mainstream readers are a serious waste of resources. A microcontroller paired with the SD card is more than enough. And CrossPoint is a gem! The list of pros is long and the device recommends itself. But there is one con that you must watch out for:

The screen is delicate.

I love cases but hate screen protectors, and tend to use my devices with care. So it has worked out fairly well for me the past few months. But you might have a different experience.

It does for me. You need to ensure that they are injected into context. Your models might be broken if they ignore it. As good as ignoring a prompt.

I subscribed to Claude Pro for one month earlier this year. The subscription lasts as long as you chat with it in the UI. The moment you use Claude Code, it becomes useless and you keep hitting limits. So, great for hammering out a specification, not so great for actually writing and debugging code.

I have a lot of hobby projects and cannot afford to drop $200/m for assistance from LLMs with above average intelligence. The Chinese models work quite well, and are cost effective, in comparison. My usage over the last few months:

- Gemini / AI Studio = translations

- Kimi 2.5 / OpenCode Go = Compiler/VM project, AI for some board games, a new document format and compiler for it,

- DS4 Flash / DeepSeek (had some credits lying around) + OpenCode Go = Template-based OCR engine

- MiMo 2.5 and MuseSpark 1.2 Contributor = The OCR engine + a product I am brainstorming + a lot of single page tools

If you select models judiciously, and can write good AGENTS.md and extremely short but potent ARCHITECTURE.md documents, the $10 you pay to OpenCode Go will last a long while even when you consume a hundreds of millions of tokens (mostly cache hits) a day. You can get multiple models to perform code review of modules against the ARCHITECTURE you have specified and get other models to execute the fixes.

I have used all three extensively. DS4 Flash is quite smart. I would rank MS 1.2 below it. MiMo is the dumbest of them all but good enough for basic stuff.

MiMo wins handsomely if you want to think about your code for minutes at a time as you write. I use it to make changes as I think. I know it will screw up some stuff. I then switch to MS/DS4 once every few hours and have it do a code review and fix the broken stuff. So much cheaper than getting MS to do it on its own.

For me, an average long session results in about 200-300M cached input, 4-800K input, 2-400K output. Mostly the lower bound. Output depends on how much the model thinks.

There are two problems here:

- cache hit pricing (both Muse Spark 1.2 Contributor and MiMo 2.5 are around the $0.002-3/M mark)

- cache persistence time

Muse Spark drops the cache in less than 5m. MiMo keeps it around for at least an hour based on my experience with whoever is serving it for OpenCode. This difference itself will inflate bills massively.

A 500K token input repeatedly read by MS 1.2 for full input price 12 times an hour = $0.60. You would be expecting $0.012. So a 50x difference. Same thing on MiMo 2.5 is $0.018 because of longer cache times.

Size is pretty similar to the original iPhone. The actual screen is the size of a credit card. You do not really notice the size once you get reading. It is light enough to be carried anywhere.

But you need a strategy to protect the screen which is the most delicate part of the entire assembly. They do sell screen protectors of various kinds and even supply one in the original box. I bought their magnetic case which works for my use case (I do not store it in a pocket with keys and other objects that might damage the screen).

I am using the Xteink X4 as my primary e-reader now-a-days. Simple ESP32 microcontroller. Plastic+glass shell with an e-ink screen. MicroSD card works like a cache. Battery lasts for weeks. Works great.

Don't get me wrong. I like modern CPUs with huge caches and incredible power, but you do not need this for most devices. Also leads to lazy programming practices and calculators consuming hundreds of megabytes of RAM.

I completely agree. IPR as a concept is suspect. But it is the world we live in, with timelines extending like crazy. There are works produced before I was born that will remain under protection till long after I am dead.

A targeted modification to the laws could produce most of the benefits for a minor cost.

Physical books and digital content is special in that you can mostly archive their content almost permanently for cheap. Buildings, paintings, idols, living things, natural features of the environment ... not so much.

So the solution is:

- mandatory copyright registration and renewal with links to where the work can be acquired

- a blanket carve out for any non-commercial trust-style org so that they can scan books etc and keep the data on their servers. They should be able to issue digital membership cards for a fee so that patrons can access the archives. Any work that is "live" based on the registration database will be locked. All "dead" material can be shared with members.

In this way, a hundred digital preservation societies can bloom.

Bought an LFP battery+inverter set recently to replace an old dying lead acid based system. We need this kind of AC-DC roundtrip systems in India because of the unreliability of grid power (random power cuts, seconds/minutes to hours long).

Chinese cells (EVE etc) and BMSes (Daly etc) are easily available and all kinds of workshops build batteries at small to medium scale for all kinds of uses: electric auto rickshaws, power backup, sink for off-grid solar. Prices are $100-200/kWh for the battery depending on V, Ah and brand.

Every couple of years, I revisit my PL dev hobby and this time I decided to create a language/runtime with pre-emptive scheduling using instruction fuel. While I always do freestanding builds, this time I decided that I also wanted to support native FFI.

That is when I realized the true horror of (g)libc. It wants to inject itself at the root of the library/program and everything from threading to dlopen/dlsym is impacted. I tried a lot of workarounds including trying to implement a loader myself, but the complexity (and fragility) grew so much that I felt it was not worth it.

Finally, I retreated into the safe world of a freestanding runtime + syscalls. FFI, if it has to happen, will occur via IPC of some kind. A second process linked against glibc that will manage calls on behalf of the clean first one.

@dang You can split this thread if it is OT

---

> Very specifically pointing out "free bus rides to women" which is very famous in India and associated with a state not run by the BJP (i.e central ruling party)

I am a pretty political/ideological person. I never try to hide that. But I support the government when I think they are right, and criticize them when they are wrong.

I guess you missed my point about cash handouts to women? That BJP governments are doing right now? My very first comment is a criticism of the government's current step! You want me to criticize the cash handouts and spare the free bus rides?

> I mean someone can say this with straight face in India today? Ffs :D

Say what with a straight face? Someone who managed to deal with the excise department but is unable to handle the GST department beggars belief. Which is what all those articles were about. Excise/sales tax/vat/GST are all the same people. Their nature is not going to change because the law changed.

> "engaging"

I am "engaging." It does not mean I have to sing the tune you expect. This is why I specifically said "But this is a different topic more suitable for a different thread," because talking about the politics of it eventually leads to flamewars and tone policing.

> article by India Today.

It is an opinion piece. About Islam on the Indian subcontinent. If the RSS is brought up, people need to know the context of their origin. About what happened on the subcontinent in that era.

Who even writes these articles?! India used to have an excise, sales tax (replaced by VAT in the early 2000s) and octroi system. And LBT (Local Body Tax) where the municipal corporation staff that do not understand the B of balance sheet could harass you endlessly on your purchase and sales record. And a hundred other random industry-specific taxes. GST replaced all of that.

Businessmen were able to navigate all this mess and were suddenly unable to handle GST? Please!

Again, we are a large country and implementation is more often than not a shit show. The bureaucracy is extremely corrupt and the politicians in charge of ministries cannot tame them. They literally say to businessmen: "The minister will leave after the election next year. I will still be here."

> But GST inspectors (many of them have seen the good old days of sales tax and VAT, and obviously don’t want to let go of their power to extract bribes) harass such firms on flimsy grounds like ‘why don’t you have a company’s nameplate clearly displayed’ that can create nuisance in residential areas. [from the Print article]

This is very common low-level harassment. very little you can do about it except rewrite the laws to make such things non-actionable.

Article 311 of the constitution makes it next to impossible to fire public servants. This is why all the corrupt/non-performing officers get transferred, not sacked![1]

[1] 311. Dismissal, removal or reduction in rank of persons employed in civil capacities under the Union or a State (https://indiankanoon.org/doc/47623/)

Yes. Habit from typing on the phone where the % sign is hidden away somewhere on my keyboard!

India is a conservative society. Women are 50% of the vote bank. They actually vote in larger numbers compared to men in many constituencies. They manage households and are very quick to notice price increases etc. The free bus rides and cash handouts are basically a transparent attempt to buy their vote. It is what it is.

> someone has to pay to maintain the infrastructure. :shrug:

Under Indian law, companies must spend 2pc of their net profit on CSR activities (corporate social responsibility). Ask banks to use that money. The Indian banking sector made a combined profit of $40B this year. 2pc of that is $800M. Close to the $1B number.

> You've got to draw the line somewhere!

Draw it in the farm sector! They do not have to pay income tax on their income, get subsidies on fertilizers, free electricity, guaranteed prices for their crops. Small farmers cannot take advantage due to average holding size, and large farmers laugh all the way to the bank!

> GST implementation

Needed. I run a business. The previous tax system was horrific in its complexity and corruption.

> Demonetisation

Implementation could have been better. But there is too much cash in the system for my comfort. Things must be digitized. We must ensure people pay their rightful share of taxes.

> unnecessary COVID restrictions etc

Hindsight. I remember how the media was blaming every single death on the government and making videos of funerals. We saw the best and worst of humanity during that period.

> He has also weakened all social welfare programs of the previous government by reducing funds or introducing restrictions designed to sabotage it for the intended beneficiaries

Have you looked at the subsidy bills? Have you looked at the finances of the state governments? You know Modi was against the rewadi culture and then had to get into competitive cash giving? I absolutely hate it. But that is the cost of doing politics in India.

Cities provide free bus rides to women, and run buses in competition to metro routes, making metros nonviable. This is such short-sighted behavior. I do not know what to say.

> fascist ideology

I do not agree with the criticism. At all. This is fairly typical of the nominal left in India and outside. Will paint with a large brush. To understand the RSS, you must go back to Hindu-Muslim relations in British India, to stoning of Hindu processions in Nagpur of the 1920s (which continue across India to this day), to the Moplah massacres and forced conversions of Hindus, and even farther back.[1]

My biggest problem with the RSS is the hubris in the top echelon and deep anti-intellectualism. Bhagwat's "same DNA" theory is a laugh riot.

But this is a different topic more suitable for a different thread.

[1] Pakistan: A 300-year-old project that Jinnah completed with Partition (https://www.indiatoday.in/opinion/story/partition-india-paki...)

> Treat the payment backbone as a public infrastructure.

Yeah. Let banks fund this out of the 2% of NP they must use for CSR activities if they have to.

> he too subscribes to the anarcho-right political ideology of transferring wealth from the poor and the middle-class to the rich

I do not see this. In fact, he is TOO socialist. He was elected on the platform (one of them at least) of "minimum government, maximum governance" and has actually continued to expand the reach of government. The bureaucracy has not been tamed. The government continues to run businesses. Build a sovereign wealth/investment fund like Singapore if you want a share of the growth in the economy instead of running loss-making businesses for decades.

And all the centralization of decision-making. He is more of Nehru/Indira than he thinks he is.

Right, this has second and third order effects. I am sure this must have been brought up during the discussions and overruled for whatever reason. The bureaucrats and politicos involved are not stupid. But there is some game being played here.