Clover-35B-A3B-Preview: a 35B diffusion language model for agent tasks
AIDiffusion Intelligence released Clover-35B-A3B-Preview, a diffusion language model with 35B total parameters and about 3B activated per token, which the source says can handle code generation and tool use. The source reports a median 1.52 seconds on 49 WindTunnel web-agent tasks versus 3.21 seconds for the Jev plus Mercury 2.5 combination, with 98.6% success against 95.9%.
AIWhy it matters.
The article explains how diffusion language models generate text in parallel and how Clover adapts them for coding and tool-using agent tasks, which helps readers compare them with autoregressive models.
Loading source posts…
Source details
QbitAI · henry · qbitai.comPublished · added here
Original title: 刚刚，扩散模型杀进Agent主场！350亿参数，速度比Jev快一倍