We evaluated GPT-6 Astra on WANDR. It scored 0.682 at $11.98 per task, the highest score of any model we tested.
GPT-6-Astra scored 13.5% higher than Fable 5.1 at 6.1% lower cost, and 27.0% higher than Opus 5 at 3.3% higher cost.
model routing — Sending each request to a model chosen by the difficulty of the task, rather than using one model for everything.
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
Why it matters
GPT-6 Astra leads Perplexity's WANDR benchmarkA standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.Full definition → at 0.682 for $11.98 per task, 13.5% above Fable 5.1 at 6.1% lower cost and 27.0% above Opus 5 at 3.3% higher cost, giving a direct capability-per-dollar comparison for model routingSending each request to a model chosen by the difficulty of the task, rather than using one model for everything.Full definition →.
postPerplexity open sources Lily, an Apple silicon engine built for hybrid agents
postPerplexity puts Claude Fable 5.1 top of its WANDR agent evaluation