Many users are excited about Fable's integrations and cost cuts such as 54% lower inference expenses because they deliver near-identical performance at much lower prices for real AI workflows.
Based on 26 visible X reactions from 44 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@omarsar0 Can confirm it works at real world scale. For my AntifragileAI platform I am implementing Karpathy’s Autoresearch methdology with cheap executor agent swarms (Qwen, Mistral etc) coordinated by a poweful planner/coordinator (e.g. Fable 5, Opus 4.8). Works like a charm.
@cognition My feedback on Devin Cloud: Used it for a complex project — performance was poor (Gemini-level bad). Burned my entire $200 plan in just 1 day. Limits vanish super fast, web UI is very buggy. Don’t use it. Don’t invest money.
@omarsar0 Fable as the orchestrator + GPT-5.6 Sol is amazing I’ve made incredible things with it Look at this galaxy travelling experience it made https://x.com/luckeyfaraday/status/2076783619004076058/video/1
@cognition Stick to first-party tools like Codex/Claude — far better experience, even with fewer features.
@omarsar0 Can confirm it works at real world scale. For my AntifragileAI platform I am implementing Karpathy’s Autoresearch methdology with cheap executor agent swarms (Qwen, Mistral etc) coordinated by a poweful planner/coordinator (e.g. Fable 5, Opus 4.8). Works like a charm.
@cognition My feedback on Devin Cloud: Used it for a complex project — performance was poor (Gemini-level bad). Burned my entire $200 plan in just 1 day. Limits vanish super fast, web UI is very buggy. Don’t use it. Don’t invest money.
@omarsar0 Fable as the orchestrator + GPT-5.6 Sol is amazing I’ve made incredible things with it Look at this galaxy travelling experience it made https://x.com/luckeyfaraday/status/2076783619004076058/video/1
@cognition Stick to first-party tools like Codex/Claude — far better experience, even with fewer features.
@cognition Well done Devin!
@cognition Amazing
Highly recommend reading. "Compared with pure Fable, Fable + Sidekick cuts cost by 54% while leaving the score nearly unchanged." Combining models has all these hidden benefits. I think this pattern might also work with GPT-5.6 Sol + GPT-5.5 (or gpt-5.6 terra/luna). https://x.com/omarsar0/status/2076724568325025920/photo/1 https://twitter.com/joon_h_lee/status/2076714221837173097
I remember when @joon_h_lee showed that Fable is cheaper than Opus in our mixed harness. We assumed it must be a bug. But then I tried it, and on one of my tasks, Fusion w/ Fable was 4x cheaper and gave a better PR. Very excited for people to try it! https://twitter.com/joon_h_lee/status/2076714221837173097
Many users are excited about Fable's integrations and cost cuts such as 54% lower inference expenses because they deliver near-identical performance at much lower prices for real AI workflows.
Based on 26 visible X reactions from 44 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@cognition Amazing
Highly recommend reading. "Compared with pure Fable, Fable + Sidekick cuts cost by 54% while leaving the score nearly unchanged." Combining models has all these hidden benefits. I think this pattern might also work with GPT-5.6 Sol + GPT-5.5 (or gpt-5.6 terra/luna). https://x.com/omarsar0/status/2076724568325025920/photo/1 https://twitter.com/joon_h_lee/status/2076714221837173097
I remember when @joon_h_lee showed that Fable is cheaper than Opus in our mixed harness. We assumed it must be a bug. But then I tried it, and on one of my tasks, Fusion w/ Fable was 4x cheaper and gave a better PR. Very excited for people to try it! https://twitter.com/joon_h_lee/status/2076714221837173097