Scott

7.7K posts

Scott banner
Scott

Scott

@scottstts

From the infinite potential of energy to the total actualization of entropy, intelligence charts a course for the pursuit of meaning, mission and love.

Dublin City, Ireland Katılım Eylül 2023
215 Takip Edilen957 Takipçiler
Sabitlenmiş Tweet
Scott
Scott@scottstts·
Introducing Threejs Awesome Graphics Agent Skills npx threejs-awesome-graphics-agent-skills install --agent codex I collected some of the best looking graphic @threejs open source projects on X and distilled them into an agent skills pack, AAA game graphics right out of the box so anyone can create games that don't look like cheap demos These are just examples included in the agent skills. All projects referenced are noted as source materials. I will continuously update this to include new projects with epic graphics github.com/scottstts/Thre…
English
16
54
502
36.8K
Scott
Scott@scottstts·
@adyshimony Well for now I think it is, maybe it’s coz in the past few days I’ve been working on mainly 3d and design, so naturally fable feels better
English
1
0
0
10
Scott
Scott@scottstts·
If I were Anthropic, I’d rather lose money to keep fable than taking it off sub Fable is still much better than Sol in hard tasks, not across the board but a solid competitive advantage Without fable tho, opus 4.8 or even 5.0 may not have a fighting chance
English
2
0
2
174
Scott
Scott@scottstts·
Finished my underwater sea park The Pearl in @threejs and WebGPU, Fable and Sol knocked it out in 2 days!! Enjoy the little montage i put together!🐟🐚 you can play it live at pearl.scottsun.io
English
2
4
40
3K
Dr Singularity
Dr Singularity@Dr_Singularity·
10T parameter models are coming. We may get our first tiny glimpse of proper AGI, possibly even ASI.
English
40
41
603
24.9K
Scott
Scott@scottstts·
@kimmonismus What makes you think b2b customers don’t care about model performance? A large chunk of Anthropic b2b customers are developers in companies, you can bet they care deeply about getting the best coding model
English
0
0
0
16
Chubby♨️
Chubby♨️@kimmonismus·
sorry, i call bs. In my opinion, Anthropic isn't worried about losing customers. And the reason is quite simple: They barely make any money in the B2C sector. Subscriptions are heavily subsidized; their compute and Fable 5 are primarily intended for businesses and enterprises, and these customers are willing to pay immensely high costs for them. This is also Anthropic's main source of revenue; it's the area where they are far ahead of OpenAI. Dario certainly isn't losing sleep over this and isn't running around hysterically because he's afraid consumers will cancel their subsidized Max plans due to the lack of Fable 5. At best, this will free up more compute for the relevant areas.
Ali Haider@ggg78g89

🚨SCOOP: MY Friend at Anthropic says things are VERY tense internally. Dario's running tough meetings — GPT-5.6 Sol is strong and Grok 4.5 is right on Opus's heels. Pulling Fable from subs on July 12 would trigger mass cancellations (why keep Max for Opus 4.8?), so they're now pushing to keep Fable 5 in subs permanently.

English
235
40
1.4K
236.8K
Scott
Scott@scottstts·
Fable one shot (400k tokens in) issues Sol couldn’t resolved in multi sessions moments like this tell me they’re not even close
English
0
0
2
74
Scott
Scott@scottstts·
Man it is so refreshing to use sol after grinding hours with fable. The speed is just amazing
English
0
0
1
117
Scott
Scott@scottstts·
@threejs Awesome Graphics Agent Skills v0.4.0 is out!🥳 npx threejs-awesome-graphics-agent-skills@latest install --agent codex added another 4 examples: - silhouette parallax occlusion @SkyeSharkie - hybrid soil/moss & procedural ivy @chirovisuals - raindrop on window (not found on X) Thank you for your beautiful creations! To check the example gallery (now has 24 awesome examples): github.com/scottstts/Thre…
English
0
0
15
270
Scott
Scott@scottstts·
so far I don’t like Sol, at least based on the past 2 days using it Feels like it’s got the slippery, manipulative, calculative trait that claude models usually have, but without the actual chop of fable I hope it can change my mind, but for now I still prefer fable
English
0
0
2
183
Scott
Scott@scottstts·
@sama You two really need to go to the ring, fight it out and be done with this whole thing for good
English
0
0
0
56
Sam Altman
Sam Altman@sama·
there are a lot of benchmarks that suggest 5.6 sol is the best model in the world right now, but the most reliable way to tell is that elon is obsessed with me again
English
3.8K
2.8K
58.7K
4.9M
Scott
Scott@scottstts·
one last ride in 7 hrs and then goodbye fable
Scott tweet media
English
1
0
1
124
Scott
Scott@scottstts·
Lost myself in underwater sea park, observatory , cable car, wonder wheel, fountains, cafe. added a lot of details, all procedural in @threejs built by fable and sol
English
0
2
15
1.5K
Chris W
Chris W@Chris_Wozniczek·
Ok so everyone had a laugh About Zuck going on X But what if muse spark is better than glm and cheaper 🤔
Ahmad Awais@MrAhmadAwais

Why is everyone ignoring Meta's new coding model? Better than GLM 5.2 yet cheaper & it has vision. Meta's new Muse Spark 1.1 model is the new underdog, being ignored by everyone. Command Code is the only coding agent where you can actually run Muse Spark 1.1 right now. We shipped it right after the launch. Meta has entered the extremely crowded AI coding market with this model, and it's pretty cheap. In the last 24hrs, there were ZERO errors on this model. I know the usage was on the lower end for us, but not that low to have no errors at all. This means, it's probably going to be a pretty nice coding model. I've been running it for a couple hours. It's legit. I actually want them to open source it. I believe Mark Zuckerberg needs a lot more push in the developer space to make this work. It's sad to see that every single partner listed on Meta's own announcement chose to ignore this launch. Cursory search, I couldn't find any. Did I miss something? None of them offer Muse Spark 1.1 integration. So what exactly are those testimonials worth? If you're going to put your name on the announcement and then ship nothing, your quote means nothing tbf. Developers in our discord community already wondering why no one else cares. We care. We've shipped it live on day one. Go use it. If Meta's leadership is reading this, partner with people who "give a shit". Zuck needs way more real push in the developer space to make this land. It's that simple. DM's open.

English
2
0
5
1.4K
Scott
Scott@scottstts·
@thsottiaux if you really want to stress test new ChatGPT super app and Sol, drop the 5 hr limit for 2 weeks
English
0
0
0
69
Scott
Scott@scottstts·
The model feels very mid I’d say, not bad but not good either. In the face of Sol, fable, grok 4.5, and glm 5.2, there really isn’t a use case for it
Artificial Analysis@ArtificialAnlys

Meta's Muse Spark 1.1 scores 51 on the Artificial Analysis Intelligence Index and is cost and token efficient compared to its peers Muse Spark 1.1 (xhigh) improves 8 points over Muse Spark 1.0 (43) in three months. It is effectively tied with GLM-5.2 (max), GPT-5.4 (xhigh), and GPT-5.6 Luna (max) at 51, three points behind Grok 4.5 (high, 54), with the leading edge at Claude Fable 5 (60), GPT-5.6 Sol (max, 59), and Claude Opus 4.8 (max, 56). The gains concentrate in Scientific Reasoning, coding, and knowledge; agentic knowledge work lags on GDPval-AA v2. @AIatMeta shared access with us ahead of public release for benchmarking. Congratulations to @AIatMeta, @finkd, and @alexandr_wang on the release! Key Takeaways: ➤ Muse Spark 1.1 gains substantially on the first Muse Spark release. This was driven in particular by gains in agentic knowledge work (GDPval-AA v2) and coding (SciCode, TerminalBench). On Humanity's Last Exam, it reaches 45%, within a point of Claude Opus 4.8 (max, 46%) and ahead of GPT-5.5 (44%) and Grok 4.5 (high, 40%) ➤ The most token-efficient of the models effectively tied at 51 and among the cheaper models to run. Muse Spark 1.1 used 94M output tokens to run the Intelligence Index, fewer than GPT-5.4 (xhigh, 109M), GPT-5.6 Luna (max, 125M), and GLM-5.2 (max, 141M). We estimate ~$0.26 per Intelligence Index task at Meta's $1.25/$4.25 pricing - below GLM-5.2 ($0.37) and roughly 3x below GPT-5.4 ($0.89) ➤ The AA-Omniscience gain is driven by abstention rather than accuracy. The score more than quadrupled from 4 to 18 as the hallucination rate fell 35 points (73% to 38%), with the attempt rate down from 95% to 82% and accuracy roughly flat (45% to 41%) Other model details: ➤ Context window: 1M tokens, up from 262k for Muse Spark 1.0 ➤ Pricing: $1.25/$4.25 per 1M input/output tokens; cache hits discounted to $0.15 per 1M ➤ Output speed: ~114 tokens/s median on Meta's first-party API, with a ~21s time to first answer token ➤ Availability: Meta's first-party API at launch

English
0
0
0
121

Keşfet

@adyshimony @jaredatjared @threejs @Dr_Singularity @kimmonismus @claudeai @ClaudeDevs @AnthropicAI