Thanks will check it out, hadn't heard of it
I use Spokenly, offline-only mode with the Nvidia model. All local, totally free. Highly recommend
Interesting, so point 2 means that a lot of the hardware being installed now won't be able to run the frontier models of 2029? How does that change the demand for compute/models in the future, I can imagine that even if…
That's a lot of effort when they could just play a short video saying 'You wouldn't steal a car' instead
Opencode Go subscription ($5 to try for one month) or Neuralwatt are what I use. Both through opensource Opencode harness (like Claude code)
It's just much more thorough and spins up a lot of subagents to basically do a lot more E2E testing. Not necessarily smarter, imo you could get the same result with a lesser model by procedurally prompting, but a lot…
Fable feels like a slightly more advanced 4.5/4.6 (less verbose than 4.7 and 4.8) with more adversial work checking. And a lot more compute to be more thorough from the first prompt. I feel it would be possible to get…
It feels to me like Fable is just a slightly more advanced Opus 4.8 (or 4.6?) but with this 'adversarial' self-challenging/checking of work and a more compute to really hunt down edge cases or to spin up many sub agents…
Even when every site in the world’s 3rd biggest economy goes down it’s still just a ‘Partial’ service disruption :D
It's a social network that became socially acceptable to browse at work. It has all the negative attributes associated with a social network and none of the upsides (apart from the occasional recruiter message).
True, found this study fascinating, basically coming to the same conclusion: https://link.springer.com/article/10.1007/s11846-020-00378-0
Nice one! Wish someone would make a browser based version a la Chronodivide's RA2
Second complete outage on railway in 2 months for us (there was also a total outage on December 16th), and many issues with stuck builds and other minor issues in the months before that. Looking to move. It's a bit of…
Great to hear but it's still the same shape. I really want a 'credit card' shaped version I can slide into my wallet.
>> the 'most significant evolution' in 50 years this is an insult to Lego Mindstorms
Thanks - I also turned it off. I guess it's a marketing thing for them, but it feels like it goes against the ethos of the company. Particularly given the fact they are clearly aware of this as they put it in the…
My Google Maps algo will be massively confused by my 'Directions' search: Sofia, Bulgaria to Kolkata, India on a Friday afternoon in January
Good for them. Makes me more likely to consider buying a Bose in future, not just because I know it won't be bricked, but also for the environmental impact of this. Kudos.
Ah thanks, didn't realise it was a 3rd party library, thought it was a claude native command
Eat 30 different types of fruits and vegetables every week. There is no 'hacking' your way to a good microbiome via these pills.
How do you manage that? /ccusage and --ccusage no longer work for me, I can only see the usage bars in /usage
50-100 PRs a week but they still can't fix the 'flickering' bug
Interesting you have this opinion yet you're using Cursor instead of Claude Code. By the same logic, you should get even better results directly using Anthropic's wrapper for their own model.
I had completely forgotten about 'Sydney' and its emoji-laden diatribes. What a crazy moment, looking back.
The 'flash' / no or low-thinking versions of those models are crazy fast. We often receive full response (not just first token) in less than 1 second via API.
Thanks will check it out, hadn't heard of it
I use Spokenly, offline-only mode with the Nvidia model. All local, totally free. Highly recommend
Interesting, so point 2 means that a lot of the hardware being installed now won't be able to run the frontier models of 2029? How does that change the demand for compute/models in the future, I can imagine that even if…
That's a lot of effort when they could just play a short video saying 'You wouldn't steal a car' instead
Opencode Go subscription ($5 to try for one month) or Neuralwatt are what I use. Both through opensource Opencode harness (like Claude code)
It's just much more thorough and spins up a lot of subagents to basically do a lot more E2E testing. Not necessarily smarter, imo you could get the same result with a lesser model by procedurally prompting, but a lot…
Fable feels like a slightly more advanced 4.5/4.6 (less verbose than 4.7 and 4.8) with more adversial work checking. And a lot more compute to be more thorough from the first prompt. I feel it would be possible to get…
It feels to me like Fable is just a slightly more advanced Opus 4.8 (or 4.6?) but with this 'adversarial' self-challenging/checking of work and a more compute to really hunt down edge cases or to spin up many sub agents…
Even when every site in the world’s 3rd biggest economy goes down it’s still just a ‘Partial’ service disruption :D
It's a social network that became socially acceptable to browse at work. It has all the negative attributes associated with a social network and none of the upsides (apart from the occasional recruiter message).
True, found this study fascinating, basically coming to the same conclusion: https://link.springer.com/article/10.1007/s11846-020-00378-0
Nice one! Wish someone would make a browser based version a la Chronodivide's RA2
Second complete outage on railway in 2 months for us (there was also a total outage on December 16th), and many issues with stuck builds and other minor issues in the months before that. Looking to move. It's a bit of…
Great to hear but it's still the same shape. I really want a 'credit card' shaped version I can slide into my wallet.
>> the 'most significant evolution' in 50 years this is an insult to Lego Mindstorms
Thanks - I also turned it off. I guess it's a marketing thing for them, but it feels like it goes against the ethos of the company. Particularly given the fact they are clearly aware of this as they put it in the…
My Google Maps algo will be massively confused by my 'Directions' search: Sofia, Bulgaria to Kolkata, India on a Friday afternoon in January
Good for them. Makes me more likely to consider buying a Bose in future, not just because I know it won't be bricked, but also for the environmental impact of this. Kudos.
Ah thanks, didn't realise it was a 3rd party library, thought it was a claude native command
Eat 30 different types of fruits and vegetables every week. There is no 'hacking' your way to a good microbiome via these pills.
How do you manage that? /ccusage and --ccusage no longer work for me, I can only see the usage bars in /usage
50-100 PRs a week but they still can't fix the 'flickering' bug
Interesting you have this opinion yet you're using Cursor instead of Claude Code. By the same logic, you should get even better results directly using Anthropic's wrapper for their own model.
I had completely forgotten about 'Sydney' and its emoji-laden diatribes. What a crazy moment, looking back.
The 'flash' / no or low-thinking versions of those models are crazy fast. We often receive full response (not just first token) in less than 1 second via API.