GPT-6 Astra tops Perplexity's WANDR benchmark at 0.682, 13.5% above Fable 5.1 at 6.1% lower cost
rohanpaul_ai · x · 2026-09-05
Perplexity published its evaluation of GPT-6 Astra on the WANDR benchmark, scoring 0.682 at $11.98 per task — the highest of any model tested.
Key numbers:
- 13.5% above Fable 5.1 while costing 6.1% less
- 27.0% above Opus 5 at just 3.3% higher cost
WANDR is unusual in testing wide-and-deep research: finding large sets of qualifying entities and backing every fact with checkable evidence. It contains 500 public tasks requiring 170,495 source-backed records, penalizing incomplete research even when found facts are correct, with strict hard scores requiring an entire branch to be right.
The result points to a substantial gain on long, evidence-heavy agentic research workloads.
Related event: GPT-6 Astra lands on Perplexity, tops WANDR benchmark(3 posts)→
More from Models
- Fable 5.5 delayed: 'Stoptober' meme says no one dares ship a SOTA model now — patience_cave · 2026-10-21
- Decision Models' routing classifier hits AntSeed, cutting inference cost to 24% of always-Opus — airesearch12 · 2026-10-21
- Developer flags anomalous Claude Projects usage: 98% weekly limit consumed at 1% thread share — 0xkarasy · 2026-10-20
- DeepSeek hands inference providers 90%+ margins, pressuring Moonshot and Zhipu hosting deals — teortaxesTex · 2026-10-20
- HelixWorld world model tested: 70GB VRAM per BF16 GPU run, consumer cards need not apply — cocktailpeanut · 2026-10-20
- Code-only Paris at golden hour: Claude's render embarrasses ChatGPT in same-prompt test — EXM7777 · 2026-10-20