Forget about the scores... look at the Cost Per Task. Fable 5.1 was already the most expensive by a good margin, and now Fable 5.2 doubles it! This is with the price cut.
The output tokens per task just exploded. From 118k tokens (46k Answer/71k Reasoning) in 5.1 to 274k tokens (81k Answer/192k Reasoning) in 5.2.
I am always interested in this comparison since it shows how quickly you can go back and forth with a model, capturing both tokens/task and tokens/second. The new models seem to be meaningfully better here.
oliwary
It's crazy how expensive Anthropic's model is compared to pretty capable DeepSeek-Flash
While they made, hands down, the best models, their 100x price for few percent better is completely made up, and I think it will end up bad for them
brachkow
That seems like an insane increase in cost for a modest intelligence increase.
But perhaps the performance on real-world tasks will be more noticeable than on benchmarks.
schnitzelstoat
Artificial Analysis is paid by AI labs and they don't publish the full method. To me it's meaningless, let's wait for other benchmarks, but if they're correct the intelligence increment is negligible.
rodorgas
https://artificialanalysis.ai/#intelligence-efficiency-tabs this is the chart that matters for most of poor developers and companies. Pareto for a given intelligence index and looks like only openai is across the quadrant with muse at the bottom. They have really cooked efficient models!
gizmodo59
8 points higher than Astra for 4.2x the cost and 5x the output tokens. Is it worth it?
Edit: high seems better than max
t0md4n
It is about time to start setting taxes for ai usage in commercial settings and set how to transition to a society with less available work. They talk so much we need universal benefits, about time for them to start paying for it
motbus3
I am wondering which use cases would require usage from Fable 5.2 especially when Opus 5.5 already performs almost all of the tasks in a very optimized way.
mutech53
To me Fable's role was always as a planner or plan reviewer, rather than something individuals or organizations would use as the primary driver. In that case, I think these improvements seem worthwhile in exchange for the cost? Ideally in that role, you want the best performing model.
Not sure if the current rise means it's cost prohibitive to use in that form too.
comments (10)
The output tokens per task just exploded. From 118k tokens (46k Answer/71k Reasoning) in 5.1 to 274k tokens (81k Answer/192k Reasoning) in 5.2.
rgbrenner
I am always interested in this comparison since it shows how quickly you can go back and forth with a model, capturing both tokens/task and tokens/second. The new models seem to be meaningfully better here.
oliwary
While they made, hands down, the best models, their 100x price for few percent better is completely made up, and I think it will end up bad for them
brachkow
But perhaps the performance on real-world tasks will be more noticeable than on benchmarks.
schnitzelstoat
rodorgas
gizmodo59
Edit: high seems better than max
t0md4n
motbus3
mutech53
Not sure if the current rise means it's cost prohibitive to use in that form too.
djacksondev