Technical performance and coding capabilities
Users are highly impressed by the model's ability to generate complex demos like GTA 6 and research data, outperforming previous Opus versions.
While developers celebrate the technical performance and coding breakthroughs of Fable 5, significant backlash exists regarding its government suspension, high pricing tiers, and intrusive safety guardrails.
Users are highly impressed by the model's ability to generate complex demos like GTA 6 and research data, outperforming previous Opus versions.
The community reacted strongly to the model's sudden removal due to US export directives and its eventual restoration following safety negotiations.
Developers express significant frustration over high subscription costs, token limits, and the perceived prioritization of enterprise clients over individuals.
Critics express skepticism regarding Anthropic's government partnerships and restrictive guardrails that sometimes block legitimate prompts or generate over-sanitized outputs.
Ok, that 3d Art Gallery build was really impressive.
Thank you, finally. Claude was getting tired of making mistakes and apologizing all the time. A toast to Fable 5!
Soon we'll be able to sit down at our computer and ask it to [Make this game], come back after breakfast, and you're playing an open world, prehistoric, sci-fi, voxel, mario kart game with your friends as the characters.
This episode is more important than ever since Anthropic pulled the Fable model.
Similar. I gave it a really hard task, basically messy code in a complex domain that was bug-ridden from a mess previously created half manually and half by Opus. It cleaned things up beautifully, both the backend and the frontend. Maybe the prompt was particularly well-suited for the model (I instructed it to put on a mathematician's hat, look at the mathematical substructure of the problem, identify invariants and general laws and verify them, then plan how to remediate). It wrote a ca. 800 line in-depth analysis (at times spawning over 130 research agents...) with remediation plans, prioritized them and then implemented them. One issue was that this document was frankly over my head. Both the language it used and the mathematical parts were very terse, and in parts it felt like a post-C2-vocab exercise. The prose was much harder to understand than the code snippets / data models. As a non-native speaker, it lost me on the prose part, and had to ask it for a less elaborate version to actually understand it. It burned the session limit four times, but it turned a huge mess of proof-of-concepts with patchy glueing into a coherent, stable application. I'm also on the Max plan using Claude Code, and I have the feeling that the harness is much more important than the consensus expectation.
"Is it worth it? Probably not". I completely disagree. That was a high quality result in the art gallery. If you are building commercial software, it's def worth it.
As another indie dev, I see my career crumbling infront of me
As a guy who studied c++ im scared
The Gta 6 build is the best game I've ever seen created by an Ai... this is insane. Im convinced that you can soon connect Ai to game engines and create real games.
If it took three years to perfect Will Smith eating spaghetti, imagine what AI will be like in 10 years time
Ok, that 3d Art Gallery build was really impressive.
Thank you, finally. Claude was getting tired of making mistakes and apologizing all the time. A toast to Fable 5!
Soon we'll be able to sit down at our computer and ask it to [Make this game], come back after breakfast, and you're playing an open world, prehistoric, sci-fi, voxel, mario kart game with your friends as the characters.
This episode is more important than ever since Anthropic pulled the Fable model.
Similar. I gave it a really hard task, basically messy code in a complex domain that was bug-ridden from a mess previously created half manually and half by Opus. It cleaned things up beautifully, both the backend and the frontend. Maybe the prompt was particularly well-suited for the model (I instructed it to put on a mathematician's hat, look at the mathematical substructure of the problem, identify invariants and general laws and verify them, then plan how to remediate). It wrote a ca. 800 line in-depth analysis (at times spawning over 130 research agents...) with remediation plans, prioritized them and then implemented them. One issue was that this document was frankly over my head. Both the language it used and the mathematical parts were very terse, and in parts it felt like a post-C2-vocab exercise. The prose was much harder to understand than the code snippets / data models. As a non-native speaker, it lost me on the prose part, and had to ask it for a less elaborate version to actually understand it. It burned the session limit four times, but it turned a huge mess of proof-of-concepts with patchy glueing into a coherent, stable application. I'm also on the Max plan using Claude Code, and I have the feeling that the harness is much more important than the consensus expectation.
"Is it worth it? Probably not". I completely disagree. That was a high quality result in the art gallery. If you are building commercial software, it's def worth it.
As another indie dev, I see my career crumbling infront of me
As a guy who studied c++ im scared
The Gta 6 build is the best game I've ever seen created by an Ai... this is insane. Im convinced that you can soon connect Ai to game engines and create real games.
If it took three years to perfect Will Smith eating spaghetti, imagine what AI will be like in 10 years time
I bet you the "safeguards" placed are just regex of security related terms.
"We're already in first place, now stop the race."
Anthropic really did this to themselves. They spent years hyping “frontier danger” as a marketing strategy. That fear narrative got them attention, and investor funding. But now that governments actually took their warnings seriously, suddenly it’s, “Oh, we didn’t mean it literally. Please release our model.” To make the irony even bigger, Anthropic sent the same executive who went to the Vatican to discuss the dangers of AI to now reassure the White House that their latest model is safe enough to release.
Oh yeah, we're going to have id verification to use fable/mythos. And then 2 months later we're going to have an open source model as powerful as mythos.
If Mythos is so great you would think Anthropic could use it to develop proper guardrails.
I wish someone instead of showing how the results look like, they show how the actual code looks like. This should be the real comparison, a beautiful website/app worth little if the underlying code is spaghetti garbage. Please include code reviews.
Funfact: I also tried and these look the same like yours. They all look the same. It takes beige/Orange very often and playfair Display. It seems every Model has its own Webdesign Template. ;-)
RTH Real Time Hallucination
This is about getting everyone to show ID to use advanced models = we're f*cked.
I think there is a serious question where the graphics assets are from since claude has no image generation. Did it buy textures? I am using Fable and it is very clear it cannot generate textures you have to supply the textures. What is the language, platform or engine used for the game? Lots of questions and it got the least time. The video spent more time on images of candles. Something is odd about that.
Graph based on sampled comments per item (n≤30)
Jaden Williams
Fireship
Pat Simmons
Anthropic
Fireship
tef
ttlequals0/claude-code-openai-wrapper
All-In Podcast
DesignCourse
Prompt Potato
Professor Of AI
The PrimeTime
SomeOrdinaryGamers
Philpax
Pragmata
bugvader
administrakt0r/shtefAIhr
rickylabs/netscript
SecOpsNews/news
PostHog/posthog
marimo-team/marimo
JoakimCarlsson/ai
GoldSH69/BlogGen
0xHoneyJar/loa-freeside
willyesposito/payroll-portal-ha
chadluo/model-watch
THEGREATCJPark/signal-ai
TheBeardedBearSAS/claude-craft
theted/vibe-chat
this-rs/project-orchestrator-frontend
this-rs/nexus
asgeirtj/system_prompts_leaks
genlayerlabs/unhardcoded
seanssoh/agent-bridge-public
darrenrolf0481-ship-it/ADHD-Sage
Akshat2634/Codelens-AI
christopherlouet/claude-base
mrgreenish/ai-blog
vikas9489/vikas9489.github.io
sjbrenchley89/Claude-skills
Anthropic12311q
wave-on-discord
pestlee
mradermacher
mradermacher
mradermacher
mradermacher
usermma
Tesleum
LeaderboardModel1
apthebest01931
mradermacher
mradermacher
mradermacher
richardyoung
micymike
micymike
Noeline
shamsghi
shamsghi
wepiqx
empero-ai
empero-ai
crosbylegal
ansulev