Leo Linsky

@leo_linsky

Autoscaling RL @ https://gertlabs.com Also: computational chemistry, quant trading, EE Generally curious

following
followers

Mercury 3 launches with claimed 1,000-plus tokens per second

The launch post claims the diffusion LLM outperforms GPT-6 Luna and Haiku 5.5 on quality.

40 Sources, 19h ago, updated 19h ago

Opus 5.5 reportedly ranks second to seventh across GBENCH coding categories

A GBENCH tester says GPT-6-Astra remains comfortably ahead, while GPT-6-Sol performs within the margin of error of Opus 5.5. The tester cautions that the effort settings likely work against Opus 5.5.

8 Sources, 26d ago, updated 26d ago

Leo Linsky Claims GPT-6-Astra Leads Multi-Agent Tests

Tweet shares results from testing in complex multi-agent coding environments.

1 Source, 45d ago, first seen 45d ago

Featured in

Mercury 3 launches with claimed 1,000-plus tokens per second

Opus 5.5 reportedly ranks second to seventh across GBENCH coding categories

Leo Linsky Claims GPT-6-Astra Leads Multi-Agent Tests

Featured in

Mercury 3 launches with claimed 1,000-plus tokens per second

Opus 5.5 reportedly ranks second to seventh across GBENCH coding categories

Leo Linsky Claims GPT-6-Astra Leads Multi-Agent Tests