In case it's not obvious, this isn't a model, this is python scripts that call out to models (code not available).
It's mostly "did you know you can attach an LLM to a PCG system?", but there is one idea here you don't see much of: an image model performs the composition (which image models are really good at), and then you extract the objects into 3d via things like SAM3D before placing them in the world, which is pretty interesting.
The rest is standard stuff you'll find in your favorite PCG system/game engine. Still, think most people don't realize how good LLMs are at 3D these days, especially Fable 5 (which this project predates).
LLMs are a PCG system. This is one of my pet peeves, I've been working on a procedural generation constraint system for quite a while (way before LLMs) as a hobby, that came with a domain specific language, where you could specify constraints.
Things like the red key must be reachable before the red door, and there should be no path that bypasses the red door, etc. written as a set of constraints.
I was working on is because I realized, constraints tend to be similar in nature (graph cut, minimum/maximum shortest path etc.) algorithmically, but its usually quite hard to control designer intent.
The reason I shelved the project, was I asked an LLM a couple years ago, to generate a dungeon room graph, following a similar set of constraints, but stated in plain English, and it did a perfect job.
Which brings me to my final point. LLMs are amazing and flexible constraint solvers, and are woefully underutilized in scenarios like this.
The trouble is the LLMs non-determinism and the "You're absolutely right, I placed the red key behind the red door rather than in front of it. This was load bearing for the whole dungeon and I blew it" moments.
With a hand authored constraint system if the red key is in the wrong place you have a bug in the constraint system that can be deterministically fixed, with a LLM you have to roll the dice on prompting again and go through expensive verification.
That’s exactly how I’ve used tools like Tripo and Meshy.
Use any image generation tool (or any image) and have a multimodal LLM like Gemini extract individual parts of the image “isolate with a transparent background” - then you can use those images for image-to-3d in those tools.
Don’t forget to use low or “smart” poly features otherwise you get too many vertices to the point you can’t performantly raycast etc.
> Don’t forget to use low or “smart” poly features otherwise you get too many vertices to the point you can’t performantly raycast etc.
I've found it's actually best to do maximal quality generation and decimate/meshopt to desired budget.
If you're raycasting against your raw art you're doing it wrong anyway, none of the big players do this precisely for the reason you mentioned. And beyond physics meshes, if you're doing anything where performance is a problem you probably want multiple LODs so you have to do this work anyway.
These days you can just ask your agent to build you a LOD pipeline and forget about it rather than sacrificing the art; Claude knows Blender.
Side tangent: Smart Poly is often a paid feature, but you used to be able to steal it for free. Because in order to preview it, you download the GLB in the Network tab.
Some 3D AI providers still leak like that, but Meshy came up with their own binary format that splits the stream up into separate files that are assembled with a response it only gets if you have Premium.
Thought it was cool technology - and wonder if browser games in general are doing stuff like that to protect 3D assets from being easily ripped from the Network tab.
It looks impressive. However open worlds are at their best with hand placed details and environmental storytelling. Look at Skyrim/Cyberpunk vs. Starfield where most of the world is proc gen’d.
I get that you run this, and then edit it. But the generated villages just aren’t interesting in my opinion. It’s probably great for tencent’s market where you’re mass producing gacha style games, but I don’t think worlds made in this fashion will scratch the open world itch like the best open world games out there.
I'd actually be curious to read an award winning ai book, all the ai books/interactive story generators I've tried have been... Ok at best... Do you know what novels have been written by ai and award winning?
Japanese isekai might not be your cup of tea, but one example is [1] which was way back in Jan, and models have improved a lot since then.
My sneaking suspicion is that there are a lot more cases than get reported, because authors are disincentivized to disclose AI usage, and readers by and large feel icky about reading AI work for reasons unrelated to quality.
I can totally see a weird future where a lot of what people like is AI generated, and everyone lies to each other about it. So it's hard to do research on this phenomenon.
Not completely surprising that AI could do well generating manga. It's very formulaic in general and by nature devoid of a lot of the relatable character development from most other writing. I have no problem reading an AI story if it's done well, but it typically just isn't. What I find annoying though is AI generated content that is only interesting if it's actually real. Like a historic photo or a video of some shocking event.
I’d be ok reading AI-written stories if it’s only “AI assisted” (I.e. there’s a skilled human driving it). If it’s just a prompt and then the AI does all the writing, I know there’s just no intention behind it as LLMs don’t actually “think”- they just (in a very sophisticated way) pattern-match against an existing corpus.
I’ve experimented using AI to prototype game ideas and inevitably once I succumb to the temptation of handing the AI the game design reigns it starts devolving to meaningless slop with a silhouette of a meaningful design. The only way to get something meaningful (that ive found) is for the human to stay in the driver’s seat.
Its a bit like saying AI is good at math when its being used by Terrance Tao. Like okay if its so good at storytelling, use it to generate a good one and share it under this comment for feedback and we can check out how good and original it is.
But that isn’t because most game writers are bad at their job, but the vast majority of the target audience don’t really care about a good story - they want to shoot the baddies and be a cool hero.
Just look what happened when a studio tries to do a bold move and have a sequel where the protagonist of the first game gets tortured and killed out of revenge, by the daughter of one of the many characters he killed in the first part. This is a direct metaphor involving the player, about grief, revenge, and loss.
They were absolutely shredded by the fans over doing this to their favourite serial killer.
1 unverified example with quite the unique, if true, circumstances against the undoubtedly countless dismissed attempts (not to mention those never submitted because they’re trash) hardly meets the bar being established here.
Even if 99% of generations are trash, why does that matter? Just throw out the failures and keep the good ones, like any writer.
Here is the text of the winning story: https://granta.com/the-serpent-in-the-grove/ If you've ever used ChatGPT to generate stories, you will find the style unmistakable. It's cloying to me, but it did manage to fool some stuffy literary judges.
In both the Autumn and Winter examples in the hero images, the algo seemed to have placed buildings on the water in the foreground
In the summer example in the hero images, the building placement + small pockets of water on the left looks odd and low attention to detail. A similar poor quality result as if an uncaring human used a scatter brush...
Curious if the examples are cherry-picked and by how much, or if this is one-shoted
I am sure they are cherry picked. No one would spend time trying to publish results like this without picking the best results they possibly could to showcase.
This kind of content generation can be quite cool to enable new game ideas for indie developers that were only possible with AAA before. But it also makes it hard to gauge the amount of human work that went into the game. When you see an elegantly designed building, environment or easter egg in the game, was that really made by a human (or even prompted by a human), or did the AI generate it autonomously as part of a much larger generation step?
There are millions of games. When choosing which one to play, you've normally chosen one that looks polished over one that does not, because of what it signals. Now that signal is gone, can't easily discern what's a sloppy crappy game vs one that's actually fun to play.
To me what matters is being able to verify human authorship at a glance. It used to be that when you saw a work of visual art you were automatically 100% sure that someone had to have spent a great deal of effort producing that, but image models took it away. This is the same situation but with game environments.
A good painting is never “just a good painting.” The context, the artist, etc. are part of the art. To remove that is to make it no longer art because it’s no longer an actual expression of anything personal. Pretty words with no intent behind them don’t mean anything.
Plenty of Holiday Inn’s have lovely painted landscapes on the walls. People aren’t paying for them to be in their homes.
With tools like this, the artist is not removed from the art. He just has more leverage.
I would argue the opposite even: this is more personal, as a singular artist can now do more by himself, without relying on others to do the parts that he couldn't do before.
AI strips a lot of the intention from the work. The artist might want a wooden house with a straw roof and while they design the house they might add framing and other small touches to the house and to the viewer the small decisions could be as important as the choice for a straw roof.
When the artist AI generates it they get a house with wooden house with a straw roof and nothing else has any thought or intention. Its less personal, the viewer previously did not have to question what the artists intention was now they do because its likely the artist had no intention or thought towards a particular part of the art.
There are a lot of assumptions here about how an artist uses the AI tools. Suddenly artists stop fussing and planning and sketching and mocking and drafting and redrafting and polishing until the work is perfect and communicates their intent in the way they need, and turn into completely different people. When a person just tosses a prompt at an LLM and accepts the result, you would be hard pressed to call them an artist.
Interesting prompt to world pipeline. I wonder if there can be more style applied than just the "MMO cartoony" style? Like medieval Witcher or realistic Cyberpunk buildings?
I'm sure the answer is yes, even now there are pretty good model generators out there that can do realistic models. Experimental things like this just tend to go with cartoony styles since mistakes and lower quality (read: faster generation) can be chalked up to "lol", whereas realistic styles tend not to be forgiving with art mismatches and shading mistakes.
In modern polished games, the art team spends a lot of time tweaking things to make sure it looks great and coherent. Most people are not willing to have their agent run for a month to do this work. I don't know if it's been tried yet but I would be really interested to see the result.
Yeah any game ready asset has to be low poly. Even if it has 4k textures the polygon (vertex) count must be low to do anything meaningful (raycast, load many models).
Usually a combination of distance and frustum culling (breaking the model down into several that only load when you need to see it) with 4k or 8k textures can make a low poly game ready asset look very high quality
The water levels don't change seasonally..So I'm guessing it's a very high rainfall area? In which case erosion and nutrient leaching are bound to be issues for the agentic farmers. I wonder if they've thought to build soil with perennial ground cover species..
But as I say this, I realise the tree species seem to change very dynamically from season to season. So I'll have to assume either the temporal rate of change is actually a lot slower than one would expect in a model like this (allowing large tree species to regenerate annually), or there are new coniferous flowering tree varieties in this universe.
Because it's a buzzword. It causes buzz. Putting it together with "world" makes it buzz even more because people want to talk about world models all the time (it doesn't matter what the actual input and output is, just that it's a world model, not some meagerly language model, it models all the world, much better than modeling just boring old language), as well as about claws. So that's why.
The project was originally called clawdbot as it was based on Claude. Pressure from Anthropic made Peter change the name to Moltbot, which everyone thought sounded gross, so he changed it again to openclaw, with a lobster mascot.
In case it's not obvious, this isn't a model, this is python scripts that call out to models (code not available).
It's mostly "did you know you can attach an LLM to a PCG system?", but there is one idea here you don't see much of: an image model performs the composition (which image models are really good at), and then you extract the objects into 3d via things like SAM3D before placing them in the world, which is pretty interesting.
The rest is standard stuff you'll find in your favorite PCG system/game engine. Still, think most people don't realize how good LLMs are at 3D these days, especially Fable 5 (which this project predates).
LLMs are a PCG system. This is one of my pet peeves, I've been working on a procedural generation constraint system for quite a while (way before LLMs) as a hobby, that came with a domain specific language, where you could specify constraints.
Things like the red key must be reachable before the red door, and there should be no path that bypasses the red door, etc. written as a set of constraints.
I was working on is because I realized, constraints tend to be similar in nature (graph cut, minimum/maximum shortest path etc.) algorithmically, but its usually quite hard to control designer intent.
The reason I shelved the project, was I asked an LLM a couple years ago, to generate a dungeon room graph, following a similar set of constraints, but stated in plain English, and it did a perfect job.
Which brings me to my final point. LLMs are amazing and flexible constraint solvers, and are woefully underutilized in scenarios like this.
The trouble is the LLMs non-determinism and the "You're absolutely right, I placed the red key behind the red door rather than in front of it. This was load bearing for the whole dungeon and I blew it" moments.
With a hand authored constraint system if the red key is in the wrong place you have a bug in the constraint system that can be deterministically fixed, with a LLM you have to roll the dice on prompting again and go through expensive verification.
Very humorous and accurate contemporary AI slop quote, thank you
That’s exactly how I’ve used tools like Tripo and Meshy.
Use any image generation tool (or any image) and have a multimodal LLM like Gemini extract individual parts of the image “isolate with a transparent background” - then you can use those images for image-to-3d in those tools.
Don’t forget to use low or “smart” poly features otherwise you get too many vertices to the point you can’t performantly raycast etc.
But yeah - it’s there.
> Don’t forget to use low or “smart” poly features otherwise you get too many vertices to the point you can’t performantly raycast etc.
I've found it's actually best to do maximal quality generation and decimate/meshopt to desired budget.
If you're raycasting against your raw art you're doing it wrong anyway, none of the big players do this precisely for the reason you mentioned. And beyond physics meshes, if you're doing anything where performance is a problem you probably want multiple LODs so you have to do this work anyway.
These days you can just ask your agent to build you a LOD pipeline and forget about it rather than sacrificing the art; Claude knows Blender.
Never had success decimating in Blender. It’s WAY too many vertices. Unusably huge.
Decimating ends up ruining the look of the model to get it down to a game ready size.
Even a simple house model, or humanoid, using Decimate in Blender there is no way that will be even recognizable.
Have you tried meshopt? It depends a lot on the art, view distance, and tuning, but it works well for my use cases.
(I agree with you about Blender though, I always hated Blender's simplify)
Have you tried meshlab [1]? I've had pretty fantastic results reducing meshes from millions to some thousands.
[1] https://www.meshlab.net
Side tangent: Smart Poly is often a paid feature, but you used to be able to steal it for free. Because in order to preview it, you download the GLB in the Network tab.
Some 3D AI providers still leak like that, but Meshy came up with their own binary format that splits the stream up into separate files that are assembled with a response it only gets if you have Premium.
Thought it was cool technology - and wonder if browser games in general are doing stuff like that to protect 3D assets from being easily ripped from the Network tab.
> Still, think most people don't realize how good LLMs are at 3D these days, especially Fable 5
Can you explain more? Can I use Fable 5 to generate a 3d model from a couple of photographs?
Again, that Claude web design aesthetic is just seemingly everywhere right now. It takes me out of what I'm reading as soon as you see the tells.
It looks impressive. However open worlds are at their best with hand placed details and environmental storytelling. Look at Skyrim/Cyberpunk vs. Starfield where most of the world is proc gen’d.
I get that you run this, and then edit it. But the generated villages just aren’t interesting in my opinion. It’s probably great for tencent’s market where you’re mass producing gacha style games, but I don’t think worlds made in this fashion will scratch the open world itch like the best open world games out there.
You know what's also good at storytelling? AI. Or else we wouldn't have writing awards being revoked when it comes out it was written by AI.
I wouldn't say we don't need artists or writers, but I think most people will be surprised how quickly fully AI driven pipelines will advance here.
I'd actually be curious to read an award winning ai book, all the ai books/interactive story generators I've tried have been... Ok at best... Do you know what novels have been written by ai and award winning?
Japanese isekai might not be your cup of tea, but one example is [1] which was way back in Jan, and models have improved a lot since then.
My sneaking suspicion is that there are a lot more cases than get reported, because authors are disincentivized to disclose AI usage, and readers by and large feel icky about reading AI work for reasons unrelated to quality.
I can totally see a weird future where a lot of what people like is AI generated, and everyone lies to each other about it. So it's hard to do research on this phenomenon.
[1] https://0115765.com/archives/165655
Not completely surprising that AI could do well generating manga. It's very formulaic in general and by nature devoid of a lot of the relatable character development from most other writing. I have no problem reading an AI story if it's done well, but it typically just isn't. What I find annoying though is AI generated content that is only interesting if it's actually real. Like a historic photo or a video of some shocking event.
I’d be ok reading AI-written stories if it’s only “AI assisted” (I.e. there’s a skilled human driving it). If it’s just a prompt and then the AI does all the writing, I know there’s just no intention behind it as LLMs don’t actually “think”- they just (in a very sophisticated way) pattern-match against an existing corpus.
I’ve experimented using AI to prototype game ideas and inevitably once I succumb to the temptation of handing the AI the game design reigns it starts devolving to meaningless slop with a silhouette of a meaningful design. The only way to get something meaningful (that ive found) is for the human to stay in the driver’s seat.
This is a great use of our global water supply
Its a bit like saying AI is good at math when its being used by Terrance Tao. Like okay if its so good at storytelling, use it to generate a good one and share it under this comment for feedback and we can check out how good and original it is.
> You know what's also good at storytelling? AI
Writing a coherent story =/= good at storytelling
It’s better than most video games. Including some of the best selling.
But that isn’t because most game writers are bad at their job, but the vast majority of the target audience don’t really care about a good story - they want to shoot the baddies and be a cool hero.
Just look what happened when a studio tries to do a bold move and have a sequel where the protagonist of the first game gets tortured and killed out of revenge, by the daughter of one of the many characters he killed in the first part. This is a direct metaphor involving the player, about grief, revenge, and loss.
They were absolutely shredded by the fans over doing this to their favourite serial killer.
Gamers by and large don’t want a good story.
What award/book was this?
The winner of a short-story prize from Granta (well known literary fiction publisher) is believed to be AI-written:
https://www.theguardian.com/books/2026/jun/20/granta-magazin...
So the author didn’t claim it was? It’s just assumed?
1 unverified example with quite the unique, if true, circumstances against the undoubtedly countless dismissed attempts (not to mention those never submitted because they’re trash) hardly meets the bar being established here.
Even if 99% of generations are trash, why does that matter? Just throw out the failures and keep the good ones, like any writer. Here is the text of the winning story: https://granta.com/the-serpent-in-the-grove/ If you've ever used ChatGPT to generate stories, you will find the style unmistakable. It's cloying to me, but it did manage to fool some stuffy literary judges.
Can you point to some aspects of the style that are typical of ChatGPT?
I have never done any story generation and after reading the first couple of paragraphs, I could not tell whether this was AI-generated or not.
If 99% generations are trash then it means it sucks at storytelling and is not good at it contary to what was claimed in the comment.
you can do both, generate a huge world and orovide aritsts with aithoring and collaboration tools to dress it. to prevent starfieldlike stuff -__-
Yeah, all that hand place details and humans do it better crap only last for like three months based on the recent past.
In both the Autumn and Winter examples in the hero images, the algo seemed to have placed buildings on the water in the foreground
In the summer example in the hero images, the building placement + small pockets of water on the left looks odd and low attention to detail. A similar poor quality result as if an uncaring human used a scatter brush...
Curious if the examples are cherry-picked and by how much, or if this is one-shoted
I am sure they are cherry picked. No one would spend time trying to publish results like this without picking the best results they possibly could to showcase.
In this competitive world, you need to integrate the cold plunge deeply with your lifestyle to stay ahead.
This kind of content generation can be quite cool to enable new game ideas for indie developers that were only possible with AAA before. But it also makes it hard to gauge the amount of human work that went into the game. When you see an elegantly designed building, environment or easter egg in the game, was that really made by a human (or even prompted by a human), or did the AI generate it autonomously as part of a much larger generation step?
> When you see an elegantly designed building ...was that really made by a human or did the AI generate it as part of a much larger generation step?
Does it matter? Why?
There are millions of games. When choosing which one to play, you've normally chosen one that looks polished over one that does not, because of what it signals. Now that signal is gone, can't easily discern what's a sloppy crappy game vs one that's actually fun to play.
To me what matters is being able to verify human authorship at a glance. It used to be that when you saw a work of visual art you were automatically 100% sure that someone had to have spent a great deal of effort producing that, but image models took it away. This is the same situation but with game environments.
Because people don't want to be tricked by AI slop.
A good painting is never “just a good painting.” The context, the artist, etc. are part of the art. To remove that is to make it no longer art because it’s no longer an actual expression of anything personal. Pretty words with no intent behind them don’t mean anything.
Plenty of Holiday Inn’s have lovely painted landscapes on the walls. People aren’t paying for them to be in their homes.
With tools like this, the artist is not removed from the art. He just has more leverage.
I would argue the opposite even: this is more personal, as a singular artist can now do more by himself, without relying on others to do the parts that he couldn't do before.
AI strips a lot of the intention from the work. The artist might want a wooden house with a straw roof and while they design the house they might add framing and other small touches to the house and to the viewer the small decisions could be as important as the choice for a straw roof.
When the artist AI generates it they get a house with wooden house with a straw roof and nothing else has any thought or intention. Its less personal, the viewer previously did not have to question what the artists intention was now they do because its likely the artist had no intention or thought towards a particular part of the art.
There are a lot of assumptions here about how an artist uses the AI tools. Suddenly artists stop fussing and planning and sketching and mocking and drafting and redrafting and polishing until the work is perfect and communicates their intent in the way they need, and turn into completely different people. When a person just tosses a prompt at an LLM and accepts the result, you would be hard pressed to call them an artist.
So the artist simply accepts the AI generated wooden house. Might it be possible that he does not?
Furthermore, before AI, artists also used assets that came from other creators, accepting them wholesale. No intention, then?
The issue I see here is that we're judging the tool, not the artist.
It reminds me so much of the digital photography backlash. But here we are.
Some people still care about it, or want a possibly shared experience over one that can only be experienced once by one person
Interesting prompt to world pipeline. I wonder if there can be more style applied than just the "MMO cartoony" style? Like medieval Witcher or realistic Cyberpunk buildings?
I'm sure the answer is yes, even now there are pretty good model generators out there that can do realistic models. Experimental things like this just tend to go with cartoony styles since mistakes and lower quality (read: faster generation) can be chalked up to "lol", whereas realistic styles tend not to be forgiving with art mismatches and shading mistakes.
In modern polished games, the art team spends a lot of time tweaking things to make sure it looks great and coherent. Most people are not willing to have their agent run for a month to do this work. I don't know if it's been tried yet but I would be really interested to see the result.
Yeah any game ready asset has to be low poly. Even if it has 4k textures the polygon (vertex) count must be low to do anything meaningful (raycast, load many models).
Usually a combination of distance and frustum culling (breaking the model down into several that only load when you need to see it) with 4k or 8k textures can make a low poly game ready asset look very high quality
The water levels don't change seasonally..So I'm guessing it's a very high rainfall area? In which case erosion and nutrient leaching are bound to be issues for the agentic farmers. I wonder if they've thought to build soil with perennial ground cover species..
But as I say this, I realise the tree species seem to change very dynamically from season to season. So I'll have to assume either the temporal rate of change is actually a lot slower than one would expect in a model like this (allowing large tree species to regenerate annually), or there are new coniferous flowering tree varieties in this universe.
Looking forward to hearing more updates from tencent 3D models. :)
impressive, but as I understand the worlds were pre generated an there no way to reproduce it on my setup?
It looks cool. Shame the source isn't available.
But Claude can do this with little prompting.
[edit] Not sure why the downvote, but Claude Design can do this, its just a matter of knowing the exact prompt.
This was created with just the prompt on from their front page and nothing else.
https://imgur.com/2pGvVcl
That's literally a cut+paste. If I had spent time on it, you can get it close to what's shown.
Claude also created that as a Blender file with assets for each of the generated items.
The assets look good, but for some reason the terrain looks too cartoonish to me.
Unfortunately, code is not available.
This looks so cool. Nice work :D
is this dwarf fortress
I kind of feel like this is how Crimson Desert was made.
It's enormous, but utterly devoid of any soul. Like eating a $3000 subway sandwich.
I hate to be an HN bikeshedder but why is everything "Claw" recently?
Because it's a buzzword. It causes buzz. Putting it together with "world" makes it buzz even more because people want to talk about world models all the time (it doesn't matter what the actual input and output is, just that it's a world model, not some meagerly language model, it models all the world, much better than modeling just boring old language), as well as about claws. So that's why.
Branding is amazing. In this case with a normal word, it's a sort of network effect?
Peter Steinberger, creator of OpenClaw, just spoke at YC Startup School. The entire thing is worth a view.
> There are 33,000 "claw" named repositories.
https://www.youtube.com/watch?v=whcfSGN6CAU&t=1450s
What does it mean in general for something to be a “claw”?
The project was originally called clawdbot as it was based on Claude. Pressure from Anthropic made Peter change the name to Moltbot, which everyone thought sounded gross, so he changed it again to openclaw, with a lobster mascot.
To me it's like a creature. A distinguishing feature of OpenClaw is SOUL.md.
It's just the natural evolution of AI into a crab.
Looks impressive, haven't looked at the code yet, I'll at night.
There is no code. The GitHub repo is empty.
"LGTM - approve"
What code, isn’t this just LLM output