Yesterday was kicking the tires on
https://helixnotes.com/, a Tauri app. It's really slick and performance is really good. Loads fast, opens quick, etc.
So, I see that as evidence that Tauri is really solid.
> top developers outside ai labs will be spending 50k USD+ on inference
I think it is more like top companies, not top developers, and the problem with developers in top companies was and is - absolute majority of them are not actually directly working on things that increase revenue, so companies can spend a ton of money and see barely if any changes in the product and the bottom line, so companies, at least legacy ones will be reluctant to sponsor that long term.
not sure how "top developers" are defined here, but there is huge diminishing return curve starts kicking in after $200/month price point for typical eng work.
I agree with the diminishing returns on spend, but worth noting that when on an Enterprise seat and paying API rates, I'd say that you can easily spend above 200/mo before seeing the curve begin to flatten
Obviously there are a ton of ways to spend money / tokens and people have different levels of experience that will put this ceiling at very different levels for different people.
I think that we are starting to see that API inference prices for US labs are excessive and subscription prices are closer to real costs so 200/m can be realistic longer term.
I mean I don't think $50k is the ceiling, unless you're talking about actual cash out. Claude code subscriptions right now can easily clear you $25-35k a year in nominal value for $2400 out of pocket cost.
Given sufficient budget and scope, I could certainly productively burn a half million dollars in tokens a year or more. I think that's where we're headed anyway, buying a 2nd or 5th claude max subscription feels slightly excessive for personal usage, but at a corporate level...
And a moderately heavy user dev can easily spend a few $K a month, so yeah. Not impossible, but a high cost, and the diminishing returns definitely kick in
I use AI to write documentation frequently. An AI agent, 600k tokens into a feature that it just authored, understands the feature as well or better than I do. It can document, compactly, what the code actually does, as well as the intent. These docs add value above having no docs and letting another AI, with zero context, read your bare code.
In the context in which I work, the quality of these docs is sufficient.
Unless your docs are being read by thousands of people, in 2026, in most contexts, handwriting docs adds less value than writing high quality instructions to your agent.
There are exceptions. For example, when kicking off a project, short crisp prose in a product brief aligns a team, removes uncertainty, and clarifies decisions. These are worth hand writing.
For some tasks, it's better. Opus refuses tasks for me pretty regularly. GLM 5.2 has never refused a task. So for anything security-related or that touches on topics that trigger Opus's safety guardrails, I use GLM 5.2.
OTOH, for anything related to UI design, I use Opus 4.8. It's much better at taking relatively vague descriptions of user interfaces and a mockup of a related UI and combining them into an immaculate design.
For anything else, I tend to run tasks in Opus and then have GLM review them and write a Markdown file with anything it finds. Then I have Opus review the markdown file and fix the issues it agrees with. The reason I usually go with Opus 4.8 first is mainly that it's faster. Opus 4.8 is, on average, about twice as fast as GLM 5.2 running on z'ai's infrastructure for the same task. There's a large variance (sometimes GLM 5.2 is pretty fast and Opus 4.8 is pretty slow), but on average it's a very noticeable difference.
When I run into Anthropic's Quota, I switch to GLM 5.2 rather than Sonnet. I don't think there's much reason to ever use Sonnet for anything if you can use GLM 5.2 instead.
This is all pretty subjective, of course. On average, I think Opus 4.8 is still a better, more reliable, and faster model, but if it went away tomorrow and I only had GLM 5.2, I wouldn't be too sad about it; I'd get things done with GLM 5.2 just fine.
The later Opus models (4.7/4.8), Sonnet 5, and particularly Fable 5 will refuse to do tasks related to offensive security.
One example I've hit is working on a benchmark of how well LLMs handle Kubernetes security tasks, there's a section on them exploiting security misconfigurations. Opus 4.6 was fine with that section, 4.7 and 4.8 saw some refusals and Fable point blank refused to do any of it.
The only other model I've seen refuse is OpenAI GPT-5.5, all the open weight models seem fine with it.
Ofc if you need to do that kind of work a lot you might be able to get on OpenAI/Anthropics allow-list for cyber work.
I’ve never had a refusal coding, and in some areas (AI red teaming specifically) I’ve found it quite good at recognizing and discussing “white hat” stuff that in the past I think would have got refusals.
But when there was the Hantavirus thing a while back, I asked it if there was a vaccine under development and got a refusal immediately. I’ve had a few like that. It seems they’ve implemented really poor guardrails on certain topics (CBRN and cyber) that have lots of false positives. But if you actually chat with the model itself it’s quite lucid about what is legitimately dangerous and what is just performative “AI Safety” style refusal.
Yeah, I’ve had Opus (and Fable) perform full security audits on my codebases that would run for 30mins. That’s what I think would have tripped it but went just fine.
Are you micromanaging your GLM costs? It seems the best bang for buck strategy right now is a Opencode Go subscription to get the subsidized rate and then switch to Openrouter's model above and beyond that + make use of a dual model strategy by having GLM 5.2 do planning and Deepseek V4 Flash for implementation.
Nice, lucky! The Opencode Go GLM 5.2 quota gets used up so fast. It's an expensive model. And while impressive for being open weight, it seems slower than Opus and GPT. So I typically only use it after exhausting quotas of discounted GPT5.5 or Opus 4.6^ paid plans.
Do you guys use it through open router? Do you have any concerns about how the data you send is being intercepted? Not that I trust Anthropic but it’s widely agreed that it’s kosher to use them for commercial work, I can’t see comfortably sending any customer data to openrouter.
Edit- I see down-thread you use z.ai directly. Same concern, aren’t you worried about using it for professional stuff.
I'm worried, but I'm worried about all of these providers. There's a good chance Anthropic and OpenAI will go bankrupt in the next five to ten years, and all of their data will go to the highest bidder.
There's no customer data sent to anyone, though. I run OpenCode and Claude Code in a Docker container that only has access to a subset of my code base. There are no secrets in there, and I'm vaguely ok with z.ai using this to train their models.
From my experience (~10M tokens on each): Opus remains better at more or less everything, and seems to hold its own better in large context work. It’s also more thorough. And usually can fix the weirdest of bugs, GLM is a bit hit and miss.
That said, GLM is worlds cheaper and a great little planner if you do a couple of rounds covering edge cases or things it might have forgotten. I can’t cache it via OpenCode Go, so I plan with GLM after gathering context cheaply, and then pass the plan to Gwen. That pairs well and shows the beauty of a multi-provider harness.
I should also add that I run GLM in Pi with a lot of cool little things such as hashline edits, some plugins for general smartness, etc. Opus runs in CC with the native tooling (except for Semble and Serena).
It's not an AI hacking into another AI's datacenter and shutting off the training. It's a slow concentration and slow alignment of power along one-party, national lines.
Nice! The world can always use a better data format.
I think you might get some traction if you post the advantages over parquet and other files directly on the readme, so that if someone goes to https://github.com/future-file-format/f3 the see why they should try it.
Mention the advantages and post metrics. Cherry pick the metrics! There's probably a good use case for this but, from the current readme, it's not clear who should use this and why.
I've been thinking about how this will play out.
I think there's a Lindy Effect. The longer the ban remains in place, the longer it's likely to remain in place.
Atherton has a *beautiful* public library adjacent to a defunct Caltrain station. It's great for kids and has a really nice cafe, ideal for VC meetings and networking.
Has anyone switched from Claude 4.7 Opus or ChatGPT 5.5 to this?
How does it feel? Dumber? Worth it for the speed? I'd love someone's subjective take on it, after doing a long session of coding.
Reiner Pope gave a talk on Dwarkesh Patel about token economics. I guess faster is a lot more expensive, generally.
Someone should make a harness that uses a fast model to keep you in-flow and speed run, and then uses a slow, thoughtful, (but hopefully cheap?) model to async check the work of the faster model. Maybe even talk directly to the faster model?
Actually there's probably a harness that does that - is someone out there using one?
I switched from Opus 4.6 -> Opus 4.7 -> GPT 5.5 and tried Flash 3.5 tonight and I was not impressed. It is straight up unreliable, e.g. deleting code and forgetting to add the new stuff it was asked to, then happily marking the task as complete with up-beat conclusion. I personally appreciate GPT 5.5 toned-down, objective style so really dislike how this model feels. I get that it's a flash model and not in the same league as GPT 5.5 but their marketing suggest otherwise so thy are just setting themselves up for disappointment.
I was using GPT 5.5 for a bunch of work this morning. It's brilliant and efficient. I was also using GPT 5.4 mini. It gets the job done and works great for subtasks that 5.5 designs. Gemini 3.5 Flash is SUCH a Gemini. It seems to work okay, but its attitude is disgusting.
"Yes, your idea is excellent."
"How this works beautifully:"
"This is a fantastic development!"
"This is an exceptionally clean and robust architecture."
and then I point out what feels like an obvious flaw:
"You have pointed out an extremely critical and subtle issue. You are absolutely 100% correct."
I'm sad that I'll probably stop using 3.5 Flash because I just hate its personality.
So, I see that as evidence that Tauri is really solid.