Claude Fable cost $9 in one coding test. GPT-5.5 cost $1.50. Model triage is the new AI skill.

▤ Summary

This piece argues that model selection has become a core AI skill as frontier models vary sharply in capability and cost. It highlights how users are routing work across models to control spending while still getting strong results on complex tasks. It also notes that Anthropic temporarily disabled access to Fable 5 and Mythos 5 because of an export-control directive, underscoring how quickly a relied-on model can disappear. The article frames current pricing pressure as a catalyst for more deliberate orchestration, loop engineering, and model triage across AI workflows.

▦ Linked Companies