DiffuSpace open-sources Clover-35B-A3B diffusion model for agents and coding

DiffuSpace, the team behind Dream 7B, has released Clover-35B-A3B-Preview, an open diffusion language model with 35B total parameters and about 3B active per token. It uses block diffusion and a MoE backbone, scoring a self-reported 73.6 on SWE-bench Verified and 51.7 on Terminal-Bench 2.1, with up to 515 tokens/s on a single H200.

• 35B total parameters, ~3B active per token, supports a thinking mode
• SWE-bench Verified 73.6 and Terminal-Bench 2.1 51.7, self-reported
• 515 tokens/s on H200 and 680 tokens/s on GB300 for agent tasks
• Weights on Hugging Face, GitHub and ModelScope under Apache 2.0