Hacker News Reader

Comment by wxw | Hacker News Reader

> Kimi Work is a Local Agent designed for deep workflows. It mounts your local folders, navigates the web autonomously via WebBridge, runs Python code in the background, and executes scheduled tasks.

It's clearly a dupe of Claude/Codex products (Codex especially, styling-wise), but I think Kimi's goal here is simply to appear on feature-parity with bigger labs. I doubt they'd want to invest much into designing a UI for "the future of agentic work" from the ground up.

Their bread and butter and claim-to-fame will continue to be low-cost, near-frontier models, for which they need to at least appear to have some app layer to plug into.

And, in some ways, duping is a great way to show people that even the app layer can be commoditized.

View on HN

I think we need to stop using the term "dupe" as if there was some originality somewhere that was not built upon or copied what already existed.

"Everything is a Remix" https://www.youtube.com/watch?v=xRHRUNpWzAY

"Good artists copy, great artists steal, the best artists make lies seem true." – Steve Jobs.

Actually, Jobs self-servingly refactored then misattributed to Picasso the line,

  "Good artists copy, great artists steal."

Jobs did seem to properly remarket Picasso's intent, though, as Picasso had said:

  "When there's anything to steal, I steal."

  "What does it mean, an artist copies? It is normal. The important thing is to do it better."

  "Art is a series of masterly thefts."

Thanks for making me smile

I love that quote and like to share it often with people around me

Love the AI format with obvious misunderstanding to this and all your other comments; makes it much easier to find AI bots like yourself.

That last line is wonderful! Thank you for this man. I honestly like the original more despite it's wordiness.

This makes me think of John Williams, an artist of great renown known for things like the Raiders of the Lost Ark OST and the Star Wars OSTs.

All of his stuff (or most of it) is repurposed and refitted chunks of other classical works. The beautiful part is, he does it so well.

Taking inspiration is one thing, this is just a knock-off. The visual language is nearly identical. The difference is in whether you are building upon something else, or just copying it wholesale and contributing nothing of your own.

Please explain where frontier models were copied from. Have LLMs always existed?

Did they invent everything that went into those LLms?

My argument is strengthened by the fact that they didn’t. There is so much technology between the raw data and the model itself. It’s absurd to suggest LLMs are simply a product of the information.

> duping is a great way to show people that even the app layer can be commoditized

Everyone is already cloning the app layer. There are 30+ claude-code clone, and codex work already cloned cowork, grok and gemini is doing the same.

I highly recommend to anyone that uses them a lot - try to clone your own. I’ve got mine fully formed into how I want it, with features I haven’t seen anywhere else yet and with integration into everything I work with. We’ve lived in a world where every app had to fit enough people well enough - but now we live in a world where an app can be molded to be absolutely perfect for just you!

This is the way. Guess what am I working right now? My own “Work” app that will be better than anything else available today!

Same here! https://github.com/rush86999/atom

Looks like the thing I am building but of course mine will be slightly different. It would be great if we could all meet to show off and discuss our AI stuff.

How did you build your own coding agent? What language/framework did you use?

This comment is a great example of how large and strange the skills gap in AI is right now.

Curious why your first impulse is not simply to point your favorite agent at a few examples and start brainstorming/planning from there?

Multiple times I’ve built a purpose specific bespoke tool starting this way. In fact, it’s a great way to learn how specialized tools are built.

The problem of the sota models clamping down on third party harnesses should be stated upfront. Getting a SOTA model in a custom harness requires API pricing or risking an account ban, correct?

This preliminary change in cost may not make it worth it for people depending on which is their "favorite agent". Especially as the default harnesses for said agents continue to improve...

> Getting a SOTA model in a custom harness requires API pricing or risking an account ban, correct?

No, only Anthropic has that policy (and I think even that is relaxed for an unknown period if you use the Claude Agent SDK: https://support.claude.com/en/articles/15036540-use-the-clau...).

OpenAI, Kimi, Qwen, GLM and Deepseek all allow it.

I'm not sure about Gemini.

> The problem of the sota models clamping down on third party harnesses should be stated upfront. Getting a SOTA model in a custom harness requires API pricing or risking an account ban, correct?

The opposite, how easy it is nowadays should be clarified. Codex's app-server quite literally is built to be integrated with, authentication is a breeze and it's quite stable, compared to the TUIs that are around.

It's a real shame Codex is moving in the direction of hiding stuff on local disks from users, hopefully they're revert the decision of encrypting the agent>sub-agent prompts so we get back introspection again...

Presumably people want to hear the opinions of other people and still believe that they might learn more from the their experience than by exclusively interacting with AI models?

The alternative of that first impulse should be what exactly? Telling AI to research Medium AI drivel articles (since nobody seems to be posting long/medium form content anywhere else these days) to figure out what’s the best approach to building an AI harness?

BTW just my personal observation but I generally find AI notoriously bad at building any AI applications compared to standard software (which is not surprising given the scarcity of high quality training data)

This was my first thought as well. I can read online review about a product but I rather ask my friends their experience with it. Same thing here.

huh? because im curious what they used? Claude Code and codex take completely different approaches. The core loop of feeding generation and having a bash tool is entirely the same. If you want to start building your own thing I'm sure there is good setups out there to start from rather than going from zero.

Ofc I can just fork gemini-cli if I want my own version but like I don't think that's what OP meant by build your own thing and customize it.

Look at pi coding agent.

Why would the language matter? Just use your favorite.

It's pretty simple nowadays if you know conceptually how they work. Running the LLM calls in a loop with tools is an agent. You only need 10 or so basic tools to accomplish nearly anything, and you can build a dynamic skill system from that. Look at https://github.com/patw/pengy, ignore the app look at the spec.md file, feed that to your current agent of choice and make your own version. Use whatever tech stack or UI you're comfortable with. Change some of the choices in how it works, so it fits what you want to work.

Implementing your own agent is very easy. Here is a minimal agent in 60 lines of Python without dependencies:

https://github.com/99991/MinimalAgent

You only need a single tool to start with. All recent LLMs know how to use bash for reading/writing/editing/executing.

Not the parent commenter, but most of it is surprisingly simple. You basically start with a "chat app" where you have a list of messages, send the whole conversation to an LLM and it replies back, which also gets added to the same list.

And you add a small twist, that instead of a 1-to-1 back and forth, you instead put it into a loop, where the LLM reply can itself "have a turn", e.g. a tool invocation, where your system is the one that replies (e.g. with the tool invocation's result). That's pretty much it, you have a 1 to potentially many "chat".

The harder part is getting all the "soft" parts right, like how to have well-behaving tool calls, timeouts, prevent huge cycles eating up tokens, but there are no one way to solve these, it's a fundamentally heuristic-heavy area.

It would have been much easier to clone six months ago but today Codex/Claude desktop apps are so feature packed it seems it wouldn’t be worth it.

For example, the mobile/remote control feature I find very useful but not sure it would be worth cloning. If you aren’t replicating every feature you aren’t exactly “cloning” it, you’ll just end up with a crappier version with a fraction of the features of the real thing.

Out of curiosity, what are your favourite features that are unique to your implementation?

Not parent, but favorite feature of my own harness I haven't seen somewhere else, is the ability for the agent to execute code on different hosts in a transparent and easy way. Fairly simple and probably could prompt the agent + set things up with ssh, but was doing other stuff related to a server<>client model, so figured why not add it as a couple of tools. Now the agent can build cross-platform applications while I'm mostly on Linux and actually verify it works, without me manually switching between three OSes.

When I tell Claude or open code using Kimi or codex to fix something on my raspberry pi, they all go to ssh right away.

I found that most things that are “features” in harnesses and tools that don’t relate to UI and UX in those harnesses themselves can be done through skills or just prompts.

A nice features of Claude’s is the remote control through the mobile app. That isn’t just a skill.

Most people don’t want to play product manager and just want to pay someone to make the decisions for them.

We used to have a time where every small business had its own custom accounting software, and discovered that these businesses aren’t actually special snowflakes and are better off just using the generic software and conforming to whatever it does

They were better off because the discomfort of shoehorning was less than the discomfort of in-house software development. The latter has changed radically so the conclusion may no longer hold. SWEs would do well to have at least one side project to calibrate their intuition about the cost of delivering and maintaining a greenfield project so that they can spot the niches which have begun to appear even as others live on borrowed time.

The accounting is very similar. The processes are not. Small businesses continue to be special snowflakes in terms of how they do their work. Both for good and bad reasons.

I’m surprised an open source accounting package hasn’t popped up that handles the accounting well. The backend could largely be the same (or at least modular) while the front legitimately needs to vary based on what type of work, how sales are done, whether they hold inventory, local taxes, etc, etc.

At least in my experience it’s usually for bad, and having to conform to standard software probably improves their process. Of course you have to pick one for the country and laws.

But dont you have to be faithful to rl environment in which it model was trained in the harness?

only labs knows what the rl enviroment was so they will always be one step ahead.

Some paradigms are simply ergonomic. Sometimes there's only one UX people will accept.

No one would buy a car steered with pedals; even if that is better (hey planes do it!) - every car maker is "copying" steering wheels and putting them on their cars. Is it controversial? If not why not?

It wasn't controversial because there was about 200 years of prior art from sailing ships.

(Before the wheel, ships were steered with tillers - which early cars also tried, interface-wise.)

Yup they have a cost advantage.

They just need to be good enough and just copy.

Sorry american labs - it’s the beginning of the end.

American labs will do fine with monthly subscriptions costing $20-300/month. They will not do fine for heavy basic LLM API calls -- it is for this that the cheap API produces substantial savings. Specialized niche provisions, e.g. excellent tool use, reliable structured output, reliable long context, cached and uncached web search, document reads, computer use, integrations, etc. also will take time for the foreign services to be good at.

you are not a trillion company if you just get 20$ a month. if that's the case, the valuation will drop 10x

I did note $20-300. In particular, domestic employers would surely still spend the upper end of it for each of their engineer staff. If there are 4 million software developers in the US of any job title, and if the employer of each spends $200 per month for the developer, that's $800M per month in revenue.

you wouldn't get the full market share, but for 1T company with 20pe you still need ~10B of revenue per month. And growing. that is hard from subscriptions alone because there are only so many developers. (google has 40B per month and is a 4T company.)

No American corporations with thousands of users will do business with foreign labs, so the American labs will remain on top.

In that case is there not some open harness that demand can coalesce around?

There are but then you're required to pay API prices for them which can end up costing significant multiples of the price of the subscription.

I am confused why they don't just offer a subscription with bring-your-own-harness, that would be a differentiator from claude/codex I think.

Negative margins to have you hooked on someone else’s software?

I am gently presuming that they are not selling tokens at a loss in their all-you-can-eat plans like the frontier models are, but maybe I'm wrong.

Actually unless they’ve changed substantially Kimi kind of does this? It’s not as advertised/focused on and the API is a little nonstandard making it hard to plug some things into, but, unlike Claude they’ll actually give you an API key you can use directly as part of your unmetered subscription. I played with it for a month or two out of curiosity, it ls not a bad value though a bit brittle and stuff

The Chinese api prices are already so cheap that that is arguably not needed.

You can also get $10/month opencode go (which officially can be used with any harness) with basically every major open weight model running inference in a place that I deem more trustworthy than the subject-to-Chinese-law inference options (but research this for yourself, there are some asterisks). This is what I use (in opencode) when I want an agent for personal use. Even if I'm being fairly intense with my usage by my standards anyway), I haven't managed to run out (yet), although I usually use the cheaper models (deepseek v4 is my favorite, and I've used both flash and pro).

From the FAQ:

> The plan is designed primarily for international users, with models hosted in the US, EU, and Singapore for stable global access. Our providers follow a zero-retention policy and do not use your data for model training.

And:

> Yes, you can use Go with any agent. Follow the setup instructions in your preferred coding agent.

Isn’t OpenCode just reselling Deepseek’s tokens over the same API? For v4-pro the prices from other providers on OpenRouter are way higher so what else could it be?

Opencode Go already seems extremely unprofitable if people use up at least 30% of their monthly tokens so its hard to imagine they’d go with several times more expensive inference provider yet show Deepseek pricing on their page.

They may have a bespoke deal with someone. They may be losing money. They may have believed a claim from Deepseek about not storing things more than they should. It may be something else. AFAIK no third party audit has been done so all that we have is what they say.

I think that pretty much all Chinese providers do that?

And Codex obviously allows that as well. Anthropic is pretty the main exception.

Vs code copilot lets you bring your own api key for services like open router. But it usually only makes sense to do this with the Chinese models because OpenAI/Google/Anthropic heavily punish you for not using their proprietary lock in harness.

So much is invested in the huge American AI companies that China undercutting them with clones that work almost as well can actually hurt the US economy.

They're at the point where copying can get people 90% of what they want anyway, the moat is drying up.

If you remove the word "AI" from your post it also describes the last 20 or so years of US/Chinese economic competition. They are real, serious competitor but also there is a lot of value in that last 10%. Some buyers need the best and you can charge a premium for it.

The speed in which a copier can catch up to a leader is faster than every before, but everything is faster these days. I think the threat is if can these labs catch up and pass American labs while keeping costs low, but I don't see that happening.

> the last 20 or so years of US/Chinese economic competition

Only with some industries, mostly electronics and light manufacturing. The rest of the history of software products doesn't look like this at all. China cloned Amazon and Meta fairly well (but only within their local/captured markets) but not Microsoft or Google and absolutely not Apple. Entertainment and culture exports are still effectively zero (like, what, three blockbuster films?) where the US, Japan and Korea define pop across the globe. Heavy industry is mixed, with decent success in consumer vehicles now but little purchase in stuff like heavy equipment or shipbuilding.

Basically it's complicated, and not really a point well suited to that kind of generalism.

Not British, but taking UK out of the definition of pop, is a bit too rough a cut. Also France. Look at the queues outside pop-glamorous stores. And was pop-art a pop artefact? Is TikTok redefining pop? Movies is an old measure, and even US has fallen with just franchise dominating. I think as a globe we are much more integrated and co-dependent, occasionally also in bad ways, than we appreciate.

I'm talking about economic activity more than the passports of the individual artists. And no, TikTok is a software product (a social media platform), not popular culture. There's absolutely popular culture being created and driven there but it's 100% NOT coming out of China. Are you high, as it were?

TikTok competes with Meta, not Olivia Rodrigo or BTS.

I am not high, I have kids, and can tell you they barely care about Olivia Rodrigo or BTS. You dispel passports as irrelevant for one leg but base your argument on them for the other. The contradiction is sufficient.

Xiaomi did a quite good job copying (and surpassing!) Apple.

They even got the car Apple always wanted, but never managed to build.

> China cloned Amazon and Meta fairly well (but only within their local/captured markets)

Americans order Chinese products from SHEIN and Aliexpress type websites.

So China did penetrate the US ecommerce market, but they are not taking on Amazon directly.

People are buying from the Chinese and reselling on Amazon, which has a good logistics network that Chinese companies can't possibly build in the USA.

> absolutely not Apple

You need to look up Oneplus to see how many people were using their devices till they announced exit from Western markets.

The Chinese Android phones are essentially replacement for a big chunk of Apple ecosystem.

If these cheap mobile devices didn't exist - Apple could possibly be selling many more devices.

> China cloned .... Meta fairly well

TikTok is giving Meta a run for their money in USA.

They had to be restricted via threat of regulation to offload stake in TikTok USA.

> Microsoft

They didn't compete with Microsoft directly as entering Western Enterprise market is hard for the Chinese. Even Google can't beat Microsoft there till now.

“You need to look up Oneplus to see how many people were using their devices till they announced exit from Western markets.”

The final part of this sentence shows why there is absolutely no need to look up Oneplus.

Oneplus is owned/backed by Oppo. Their other phone brands like Oppo, Realme, etc. are really popular worldwide. They removed Oneplus from some regions (e.g. EU) where some of their other brands are more popular.

I'm not sure if you are from the US, but if you are, it really gives you a distorted image of what the worldwide smarphone market is like.

E.g. in many European countries, it is 1/3 iPhone, 1/3 Samsung, 1/3 others (which are mostly Chinese brands).

Seems like you are not aware of the situation.

To get a better idea on the topic that you commented on and views of members of HN

You can read comments on this post - https://news.ycombinator.com/item?id=48932539

> Entertainment and culture exports are still effectively zero (like, what, three blockbuster films?) where the US, Japan and Korea define pop across the globe.

And this is really sad, because they have thousands of years of cultural heritage and also plenty of great films. Somehow, they've failed at exporting this completely.

Well, another way to put it is that the western markets managed to keep it out via propaganda and other soft influences on people.

> Heavy industry is mixed, with decent success in consumer vehicles now but little purchase in stuff like heavy equipment or shipbuilding

Uhh, China does over half of the world’s shipbuilding.

China is also responsible for 25% [1] of the world’s heavy equipment manufacturing and is considered to have double[2] the CAGR of the global average for heavy equipment manufacturing.

0: https://www.csis.org/analysis/china-dominates-shipbuilding-i...

1: https://www.grandviewresearch.com/horizon/outlook/constructi...

2: https://www.grandviewresearch.com/industry-analysis/heavy-co...

the difference is that the US economy is currently over-investing in AI to the point where it's a significant portion of the economy, it's the biggest competitive opportunity in quite a while

>Some buyers need the best and you can charge a premium for it.

Yes, and that is how you get a top-heavy k-shaped economy I believe. It is harder and harder to sell premium A+ products, because most people can only afford what works. If my old ice cream vendor charges me 25% more but puts chocolate sprinkles on it i'm going for the non-sprinkle version.

China isn't undercutting anything. Kimi has to be hosted somewhere and is extraordinarily expensive to operate at Fable-levels of capabilities. The moat is (and will always be) the cost to actually operate these models, and that cost will continue to soar, it will not reverse, and the hardware will never keep up such that at-home use somehow becomes economical.

It's like releasing a cutting edge search engine, proclaimed to be on par with Google. Oh yeah, I forgot to mention, you need to actually run it somewhere. One of the core moats for Google has been the cost to scale and compete globally with a search engine, which only a few entities on earth could even attempt. Somehow the HN crowd forgets this every single time. The same claims were endlessly made regarding Uber, as though the app was the hardest part of competing with Uber. Here's a hint: the model isn't the only super hard part of the total process at this point. Try running something akin to GPT 5.6 for a billion users, see how much that costs.

The winners will be the owners of the subscriptions and API usage. That's going to be OpenAI and Anthropic as of now, possibly Google will stay in the fight with their scale advantage focused on the consumer. It will never be the open weight models out of China. Who is going to invest the trillion dollars into infrastructure needed to run the latest version of Kimi 3-5 years out, for a billion users? There's only one entity that would ever consider it, and that's China domestic.

And the real money (for a sustainable business model allowing for constant massive reinvestment & infrastructure maintenance) is in pairing up advertising, which is another spot where the open weight models from China will never compete. It doesn't matter if the HN crowd hates this outcome, nothing will stop it from being the future core of mass consumer usage. And the business use will stay with subscription + API use as well: the best option for that today is OpenAI and Anthropic.

If you think Open AI can make hundreds of billions of dollars selling subscriptions for a chatbot I have a bridge you may be interested in.

>Who is going to invest the trillion dollars into infrastructure needed to run the latest version of Kimi 3-5 years out

Everybody that is concerned about their data ownership and residency, and how the US political system is behaving right now. There is a huge incentive to cut off ties with API and subscription providers, especially those based in the US.

> It's clearly a dupe of Claude/Codex products

Have you seen Kimi Code ?

It's hard to believe it's a different code than Claude Code.

I want to believe I work on something creative in this space (even though, the current UI was never "meant to be" and (d?)evolved from a substrate of cheaply available UI components and laziness). THen again, much of the unorthodoxy actually comes from all the small(and big) usability nightmares that I can not bring myself to tolerate while working in a desktop environment(so, created new ones - like bitmap-based context trees ;)

Copying and reinterpreting/reimplementing successful technologies and processes is the bedrock of human civilization

> Their bread and butter and claim-to-fame will continue to be low-cost, near-frontier models, ...

K3 is beating Fable on several benchmarks. So I take it you don't consider Fable 5 to be frontier.

What are the frontier models?

People endlessly claimed the same thing about DeepSeek upon its release. they claim it every single time a very large cutting edge model is released out of China. It's a combination of propaganda, some truth, and the faux open culture desperation to see big tech get destroyed (somehow better if big China takes its place instead).

https://arena.ai/leaderboard/code

(that's just one, there are others... and several are claiming that those saying "yes sure it's just as good, but it uses more tokens" ain't really true either).

And I don't think anyone is excited about China releasing anything: I think people like the idea of open-weights.

weird to all it a dupe, these are new paradigms, it's like saying BMW duped Mercedes Benz cause they both make stuff on 4 wheels

Reply on news.ycombinator.com