Shawn Fumo

[Shawn Fumo @ShawnFumo]

@TinyThoughs @anakin @ClaudeDevs They could have handled all of this better, but if it is essential apps for your business, just pay by-the-token to begin with. That's what it actually costs. The monthly plans are just a crazy discount. Also, don't have to use Opus xhigh for every task.

[Shawn Fumo @ShawnFumo]

@Peakanime0 https://t.co/RCgRURCW76

[Shawn Fumo @ShawnFumo]

@darkseidzz @logangraham @AnthropicAI OAI did this too with the math olympiad model. We don't even know if that was incorporated into GPT5.5 or not.

[Shawn Fumo @ShawnFumo]

@OneSpicyMeatBol @ClaudeDevs The monthly plans are purely a promotional loss leader. They were trying to treat it as a gym membership, but then everyone started showing up to the gym and it all fell apart. We use Bedrock and there's never quotas, it is never down, etc.

[Shawn Fumo @ShawnFumo]

@OneSpicyMeatBol @ClaudeDevs I know it's annoying that they keep changing stuff up. But the real answer is if you're using it seriously for business, just use the per-token price, either directly with them or through Bedrock on AWS. Yes that's expensive, but that's what it actually costs.

[Shawn Fumo @ShawnFumo]

@yadnesh_sa88965 @ClaudeDevs You probably should optimize costs at least a bit. Use Opus just for planning or difficult spots. Use Sonnet to implement. Use Explore sub-agent (which uses Haiku). Don't just leave it on Opus 4.7 xhigh for everything. It'll make a big difference.

[Shawn Fumo @ShawnFumo]

@jlin1206 @jeffdafo @yadnesh_sa88965 @ClaudeDevs We use Bedrock and work pays for tokens, but there's plenty of ways to optimize if you need to. Plan with Opus, then implement with Sonnet. Make sure research is with the Explore sub-agent which uses Haiku, etc. If you're using Opus 4.7 xhigh for all, of course it'll cost a lot

[Shawn Fumo @ShawnFumo]

@GanzaloraX @ReadySettJet @ClaudeDevs You could use opus for planning and sonnet for implementation. That plus emphasizing using the explore subagent (which uses Haiku) would lower costs a lot.

[Shawn Fumo @ShawnFumo]

@mlitwiniuk @theo So this wouldn't be the first time that there was way too simple of a grep somewhere.

[Shawn Fumo @ShawnFumo]

@mlitwiniuk @theo At least this one has the better error message of third-party apps. The detection might just be them implementing it poorly. I saw in the changelog that for a while it was reporting github (through gh) had a rate limit error if your commit message mentioned rate limits.

[Shawn Fumo @ShawnFumo]

@DebatableChild @CuiMao @DarioAmodei The fact that the text is wrong makes me think it is probably text to video. If it was just in-painting, it'd probably leave that alone. Plus if you notice, the two people walking outside at the end are waiters. So it's like it wanted to show them but put them in the wrong spot.

[Shawn Fumo @ShawnFumo]

@brushfushstuff @CuiMao @DarioAmodei Yeah, exactly my process too. "What do they mean by transit station? Oh they said it is AI". Then you notice the text, then the waiters on the outside. But the zoom happens before you notice the text, and don't notice the waiters at first if you're distracted.

[Shawn Fumo @ShawnFumo]

@GFaang97609 @cixliv It's definitely not AI, if you look on YouTube. Is a long video and very high res, frame rate, consistent. If anything, is CGI, but they've pulled off a lot that people was fake in the past, so I dunno.

[Shawn Fumo @ShawnFumo]

@somalirev @karpathy But asking for one word, it should know better that it has a "disability" for seeing individual characters and should double-check, but it is a bit "overconfident" and answers off the cuff and sometimes gets it right and sometimes wrong.

[Shawn Fumo @ShawnFumo]

@somalirev @karpathy I think some of the issue with this tends to be whether it "thinks" or not. It's similar to the counting letters issue. If you ask it to count the letters in a sentence, it is more likely to get it right. It's like "oh, that's hard, let me use code to do it".

[Shawn Fumo @ShawnFumo]

@thales_comlb26 @karpathy I knew someone who had a business extracting text from pdfs. He said depending on the particular files, you'd have to fall back a halfway OCR to try to figure out the logical groupings of the text. Like figuring out if text in the same "line" was one paragraph or two columns.

[Shawn Fumo @ShawnFumo]

@thales_comlb26 @karpathy Though (on a slight tangent), I've heard PDFs tend to be a lot tricker than one would assume, since they were developed more for printing than for documents originally. Like the letters could be output in the order they were typed instead of semantically.

[Shawn Fumo @ShawnFumo]

@RyanJTopps @karpathy And honestly Sonnet 4.6 is good for quite a lot as well. Usually I switch to Opus when something gets stuck vs defaulting to it.

[Shawn Fumo @ShawnFumo]

@Du8TGKveKhp1MMb @interesting_aIl It is purely about flexibility, certainly nothing to do with toes. Almost anyone can do it if they gain enough flexibility especially in their hips. Proportions (leg vs body length) can make it a bit easier/harder, but there's leeway there too.

[Shawn Fumo @ShawnFumo]

@adam_adair_ @DaisyGray2027 @interesting_aIl For me, I found it it was more about hips than ankles. Doing stuff like seated good mornings and frog walks stretched the hips more.