GPT-6 Astra leads every comparable benchmark against Claude Fable 5.1, Claude Fable 5, and Claude Opus 5.
Clear wins (large gaps)
ARC-AGI-3: Astra 98.6% vs Opus 5 30.2% (high). No scores for the Fable models.
FrontierMath Tier 4 (v2): Astra 97.6% vs Fable 5.1 / Fable 5 87.8% and Opus 5 73.2%.
AutomationBench: Astra 41.4% vs Fable 5.1 31.4%, Opus 5 26.9%, Fable 5 17.4%.
Terminal-Bench Science 0.1: Astra 64.6% vs Fable 5.1 52.6%, Opus 5 29.0%, Fable 5 24.7%.
BenchCAD: Astra 95.9% vs Fable 5.1 84.3%, Opus 5 82.1%, Fable 5 67.5%.
ExploitBench: Astra 100% vs Opus 5 70%. No Fable scores.
Agents’ Last Exam: Astra 59.3% vs Opus 5 52.7% and Fable 5 48.7% (xhigh). No Fable 5.1 score.
Closer but still ahead
DeepSWE v1.1: Astra 74.1% vs Fable 5 69.9%, Opus 5 68.8%, Fable 5.1 67.4%.
GPQA Diamond: Astra 96.0% vs Fable 5.1 93.7%, Opus 5 93.2%, Fable 5 92.6%.
HealthBench Professional (length-adjusted): Astra 63.4% vs Fable 5 60.9%, Opus 5 57.5%, Fable 5.1 56.6%.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
wow.
I wonder how many parameters this beast is.
Almost certainly will still be too expensive for regular use. Still this points to the fact that a lot of potential growth still exists for llms and with so much competition the future will bring lower prices.