Im bad at keeping score of model lineage and i generally just ignore the rage posts on Reddit where someone is grieving the loss of his prior bestie model to the latest release
So when openAI or Anthropic pimp their new model to me I sheep it up. But my pattern is to use both (often for entertainment feeding them the same prompt) and generally I prefer Claude for design and complex code stuff while ChatGPT is my goto for more conversational stuff and brainstorming (including writing)
But lately Claude has really annoyed me to the point where if it wasn’t so *(#& good at coding I would’ve cancelled it and called ChatGPT good enough. I never really put my finger on it but it was def vibes. It just gave me the ick a lot
Reading this I’ve conscientiously switched to 5.5 and life is better
Agreed 100%. For any casual chats or "hey, can you check this for me," ChatGPT is miles ahead. But Opus 5.5 is finally, finally not awful and is growing on me. Happy to hear you feel the same!
Fair. You're right to push back on that. The proof of the comment was in the doing. Communication efforts demand concision. It's not in the words that meaning solidifies. [Continue like this for several paragraphs at least.]
Surprised you stuck with Opus 4.6 for so long. As soon as I tried Fable 5, I was hooked. Opus 5 felt awkward from the beginning. I'm using Fable 5.1 now and haven't taken the plunge on Opus 5.5 yet.
I tried Fable 5 when it was briefly available on subscription and was quite happy with it, but now that it requires usage credits, it's out of my reach. But Opus 5.5 is growing on me and it's allegedly better than Fable 5.1, so I'd be curious to hear how they compare in your experience if you end up trying it!
Daniel, I have the $100 Max plan and can use Fable without usage credits. I use a delegation protocol when agents are used to keep Fable as the orchestrator and send out lesser models to do the agent work. I haven't hit a session limit since, unless I am processing lots of articles. I do research which is less intense than coding, though.
Ah, then I get why Fable's a go-to for you! I'm just on the Pro plan and have ChatGPT Plus as well, so I switch between them and it's more than enough for my needs right now. Have you given Opus 5.5 a shot and tried comparing it to Fable?
I don't typically use LLMs for big life decisions this way, but it's curious to observe the difference between how these models handle such tests, so this was s fun thing to try. Do you normally use LLMs for life stuff?
I suppose it depends on how you view life. Today, I used Gemini to create a revised exercise plan for my gym workouts. What took (with my prompting) about an hour to create a complex yet usable plan would have been many hours/days without Gemini.
I am also learning Garage Band (a DAW) to support my singing and modest composing objectives. I integrated Gemini's perspective with the course, which I find to be a powerful combination for me. I can ask questions about topics for which I need more information or nuance. There is very little in my life that a LLM can't improve, if only to get the weather forecast and ask questions if I want more detail. :) Good to talk with you.
Ah, in the case, yeah, that's the same for me! I do lots of day to day research and learning through LLMs, and Claude built me a full-fledged goalie tracking app for my son's hockey, etc.
Agreed that there are so many ways to use an LLM for something fun and productive if you're willing to try!
With Garage Band, it would be fun to try and see if ChatGPT or Claude with their browser use could successfully use the software for you to demo stuff.
First, let me say I enjoy our conversations and always learn something. Second, I think your suggestion to combine GB with an LLM is BRILLIANT. I would like to pursue that with you; perhaps we could collaborate. Can we discuss when you have some time?
As to music, I have been an enthusiastic amateur singer since high school, when I played banjo and guitar in the folk era. I sang in my college choir and then discovered barbershop quartet singing, which I enjoy to this day. I have also sung in church choirs and choruses of varying shades. Basically, I am a music junkie. :) I use AI in many ways, for example, to identify vowel sounds and how to interpret songs. Talk soon.
Im bad at keeping score of model lineage and i generally just ignore the rage posts on Reddit where someone is grieving the loss of his prior bestie model to the latest release
So when openAI or Anthropic pimp their new model to me I sheep it up. But my pattern is to use both (often for entertainment feeding them the same prompt) and generally I prefer Claude for design and complex code stuff while ChatGPT is my goto for more conversational stuff and brainstorming (including writing)
But lately Claude has really annoyed me to the point where if it wasn’t so *(#& good at coding I would’ve cancelled it and called ChatGPT good enough. I never really put my finger on it but it was def vibes. It just gave me the ick a lot
Reading this I’ve conscientiously switched to 5.5 and life is better
Agreed 100%. For any casual chats or "hey, can you check this for me," ChatGPT is miles ahead. But Opus 5.5 is finally, finally not awful and is growing on me. Happy to hear you feel the same!
Remind me NEVER to try Opus 5.0.
Here's your reminder to NEVER EVER try Opus 5. You're welcome!
Whoa, that's a little much, don't you think?
Fair. You're right to push back on that. The proof of the comment was in the doing. Communication efforts demand concision. It's not in the words that meaning solidifies. [Continue like this for several paragraphs at least.]
You're absolutely right! Having reconsidered my initial position, I can see that I was too hasty.
Surprised you stuck with Opus 4.6 for so long. As soon as I tried Fable 5, I was hooked. Opus 5 felt awkward from the beginning. I'm using Fable 5.1 now and haven't taken the plunge on Opus 5.5 yet.
I tried Fable 5 when it was briefly available on subscription and was quite happy with it, but now that it requires usage credits, it's out of my reach. But Opus 5.5 is growing on me and it's allegedly better than Fable 5.1, so I'd be curious to hear how they compare in your experience if you end up trying it!
Daniel, I have the $100 Max plan and can use Fable without usage credits. I use a delegation protocol when agents are used to keep Fable as the orchestrator and send out lesser models to do the agent work. I haven't hit a session limit since, unless I am processing lots of articles. I do research which is less intense than coding, though.
Ah, then I get why Fable's a go-to for you! I'm just on the Pro plan and have ChatGPT Plus as well, so I switch between them and it's more than enough for my needs right now. Have you given Opus 5.5 a shot and tried comparing it to Fable?
I appreciate using a powerful LLM to address day-to-day issues. This reminds me of my therapy days, and the power of non-judgemental listening.
I don't typically use LLMs for big life decisions this way, but it's curious to observe the difference between how these models handle such tests, so this was s fun thing to try. Do you normally use LLMs for life stuff?
I suppose it depends on how you view life. Today, I used Gemini to create a revised exercise plan for my gym workouts. What took (with my prompting) about an hour to create a complex yet usable plan would have been many hours/days without Gemini.
I am also learning Garage Band (a DAW) to support my singing and modest composing objectives. I integrated Gemini's perspective with the course, which I find to be a powerful combination for me. I can ask questions about topics for which I need more information or nuance. There is very little in my life that a LLM can't improve, if only to get the weather forecast and ask questions if I want more detail. :) Good to talk with you.
Ah, in the case, yeah, that's the same for me! I do lots of day to day research and learning through LLMs, and Claude built me a full-fledged goalie tracking app for my son's hockey, etc.
Agreed that there are so many ways to use an LLM for something fun and productive if you're willing to try!
With Garage Band, it would be fun to try and see if ChatGPT or Claude with their browser use could successfully use the software for you to demo stuff.
What kind of singing do you do?
First, let me say I enjoy our conversations and always learn something. Second, I think your suggestion to combine GB with an LLM is BRILLIANT. I would like to pursue that with you; perhaps we could collaborate. Can we discuss when you have some time?
As to music, I have been an enthusiastic amateur singer since high school, when I played banjo and guitar in the folk era. I sang in my college choir and then discovered barbershop quartet singing, which I enjoy to this day. I have also sung in church choirs and choruses of varying shades. Basically, I am a music junkie. :) I use AI in many ways, for example, to identify vowel sounds and how to interpret songs. Talk soon.
Heya, what I meant by letting LLMs use GB is to literally let them press buttons, arrange tracks, etc. using "Computer Use": https://www.whytryai.com/p/when-computer-use-useful
Would be curious to hear if that's something they manage competently enough from your perspective!
But lemme know about the kind of collaboration you're thinking about.
And quite a music backstory! Nice to see that you find AI helpful in the process rather than something soulless and counterproductive.
Thanks very much. I will review, but I think the concept is exciting. I will experiment and report back. Have a good weekend.