On Friday I wrote that Opus 5.5 does my work and a Fable 5.2 agent checks it, because Fable is too token hungry for everything. A new benchmark backs that up.Deepnote's Agent Initiative Benchmark tested 12 models as proactive agents in the Slack of a simulated company: help when needed, stay quiet when not.Fable 5.2 scores 4% higher than Opus 5.5, but makes twice the tool calls in twice the time, at $1.87 a task against $1.28. OpenAI's GPT-6 Astra and GPT-6.1 Sol top the list, Sol at $0.11 a task.The benchmark, dataset, and code are open source:https://deepnote.com/research/benchmark#AIAgents #ClaudeCode #LLM
Related
@Heliograph i know right :awesome: and there's more 😆 but this instance only allows #500chars so you have to go and read...
@Heliograph i know right :awesome: and there's more 😆 but this instance only allows #500chars so you have to go and read it yourself on that ancient #globaleconomy #accounting #exp...
$20 - 30 is all it takes to get a powerful AI video ad for your small business! 🚀🎬 Alov AI handles the heavy lifting:•...
$20 - 30 is all it takes to get a powerful AI video ad for your small business! 🚀🎬 Alov AI handles the heavy lifting:• Your idea• Our professional script, soundtrack & AI visuals...
@eniko Sometimes the markets are very slow to respond. Especially when the major players are hopped up on ketamine and s...
@eniko Sometimes the markets are very slow to respond. Especially when the major players are hopped up on ketamine and shrooms and sitting around the table waiting for somebody to ...