B. Salmon

[B. Salmon @salmon38619]

@xsphi Most people don't want to learn. I'm in finance where people use Excel all day for years and they can barely do a VLOOKUP or pivot table, let alone Power Query. You think they're going to learn how to use Codex?

[B. Salmon @salmon38619]

@theo @theo - can you do a response vid? I love your content btw keep it up 😀

[B. Salmon @salmon38619]

@rezoundous Haha already cancelled it.
Opus 5.5 all the way

[B. Salmon @salmon38619]

@TokenGremlin I find it shocking that they've done this and haven't announced anything about it yet. If I was considering upgrading to the $100 a month plan, I wouldn't because I just don't understand how much I'm going to get.

[B. Salmon @salmon38619]

@ItsCuthulhu @0xSero I love that picture, it's brilliant. We're prisoners of our own minds.

[B. Salmon @salmon38619]

@SilasHelyfts @scottstts Why do you say that Claude chat is not token efficient?

[B. Salmon @salmon38619]

@googledevs @GoogleAIStudio It's very good, but unfortunately, you're on tier one billing when you pay money, which restricts you to 100 requests per day. So you have to be really conscious about how you use it. Otherwise, you can blow through your daily budget by midday.

[B. Salmon @salmon38619]

@GoogleAI @GeminiApp @Android @GoogleAIStudio @antigravity @googlechrome I've been using it recently and it's brilliant. I'm just curious what app were you using when you just dictated? I've never seen that UI element in the bottom center before.

[B. Salmon @salmon38619]

@get_utter @theo @joshtriedcoding I've used GPT Transcribe and I found it very good. Do you think that's the best out of those three?

I'm willing to try the others.

[B. Salmon @salmon38619]

@Macrodose1 @theo @joshtriedcoding @openwhispr Local models suck compared to Whispr Flow and Typeless, which do both transcription and cleanup. You can do both of those as well in openwhispr, but the quality is just not as good and it's much slower.

[B. Salmon @salmon38619]

@theo @joshtriedcoding I've been using Typeless the last couple of days, and I do prefer it to Whispr Flow. But I've just had a look at the pricing per month, and it's $30 a month. No fucking way am I paying that much a month for dictation.

[B. Salmon @salmon38619]

@astropol0 Copium.

[B. Salmon @salmon38619]

@thomasboomcom @theo @typelessdotcom @WisprFlow Thanks for the recommendation. I've just downloaded it and I'm going to test it now. Just went through the tutorial, it was pretty impressive. A bit slower than Wispr Flow, but the clean-up is vastly superior.

[B. Salmon @salmon38619]

@CtrlAltDwayne They also double the tokens per second.

[B. Salmon @salmon38619]

@05H_31 @dummerspast39 Tibo said that the pricing change will benefit subs.

[B. Salmon @salmon38619]

@TrueOne1776r @HarshithLucky3 No, they share the same pool like SuperGrok.

[B. Salmon @salmon38619]

@claudeai Does that mean that our subscriptions will go further now that it costs 40% less to run?

[B. Salmon @salmon38619]

@melvynx You forgot to mention it also takes much longer per task as well. It's basically a downgrade from 4.6. No point in using this model compared to its competitors which are cheaper, faster and similar in intelligence.

[B. Salmon @salmon38619]

@DavidOndrej1 fast....