Crazy stuff. #Claude #Fable5 pulled off an exploit last night optimizing my Chinese token costs. It was already inexpensive, but #Anthropic's AI spontaneously found a solution to reduce the cost by another 5 times by brilliantly leveraging prompt technology.
PROMPT ENGINEERING
-
Bernard Marr: Talking to AI a Top Business Skill for Better Results
By
–
How To Talk To AI #Talking to #AI well is becoming one of the most valuable #business #skills of the decade, because better questions lead to better #thinking, clearer #decisions and stronger results. Jamie Bartlett, author of How To Talk To AI, explains why the real
-
A good prompt is a clear spec; writing them is the challenge.
By
–
A good prompt is just a clear spec, and writing those was always where people struggled.
-
Fable’s long tasks create a Claudish dialect, need plain English
By
–
One thing I mentioned only in passing in my Fable post is that, for long running tasks, Fable starts to develop its own dialect as its many agents and tasks reinforce themselves and make Claudish language ever more Claudish. You need to ask it to report out in plain English.
-
Claude Fable 5 Launch with Auto-Routing to Best Models
By
–
Claude Fable 5 Launched On ChatLLM Try it for hard-coding prompts!! Auto-route to the best model based on prompt hard-coding – Fable 5
design – Opus 4.8
video – Grok Imagine 4.5
image – GPT Image 2 Our agent will automatically find the best AI for your task -
Goal with precise prompt automatically executes
By
–
Just set a /goal with a chonky well defined prompt and it just goes off and rips
-
Claude Fable consistently nails harder tasks with /loop prompts
By
–
I keep throwing harder and harder tasks at Claude Fable and it just keeps nailing them. I’m genuinely struggling to find something it doesn’t get on the first try with a good /loop prompt.
-
Fable’s pelicans for thinking effort levels and API costs
By
–
Here are Fable's pelicans for the different thinking effort levels, plus how much each one cost to generate via the Claude API