Instruction following
Agentic IF Index (Internal)
Meta's internal measure of whether a model keeps following the instructions it was given while working as an agent — over a long run of tool calls, not just in a single reply. Higher is better.
Rankings
Higher is betterAgentic IF Index (Internal) — frequently asked questions
- What is Agentic IF Index (Internal)?
- Meta's internal measure of whether a model keeps following the instructions it was given while working as an agent — over a long run of tool calls, not just in a single reply. Higher is better.
- Which AI model scores highest on Agentic IF Index (Internal)?
- Muse Spark 1.3 by Meta holds the best Agentic IF Index (Internal) result among tracked models, at 57.8% (released Sep 2 2026). Higher scores are better on this benchmark.
- What are the top 1 models on Agentic IF Index (Internal)?
- 1. Muse Spark 1.3 (Meta) — 57.8%.
- How many models have a published Agentic IF Index (Internal) score?
- 1 tracked model has a published Agentic IF Index (Internal) score. Scores are the figures reported by each lab at that model's release, so this page is a record of results over time rather than a re-run leaderboard.