DiffuSpace, founded by the HKU Dream team, open-sources Clover-35B-A3B, a diffusion LLM hitting 680 tokens/s on one GPU and 73.6 on SWE-bench Verified
On Oct 28 DiffuSpace, whose members come from the HKU NLP group behind Dream 7B, released its first model, Clover-35B-A3B-Preview: a mixture-of-experts diffusion language model initialized from Qwen3.6-35B-A3B, with 35B total and about 3B active parameters, a 128K context window and block-parallel denoising. Vendor-reported results include 73.6 on SWE-bench Verified, 51.7 on Terminal-Bench 2.1 and 83.8 on GPQA Diamond, with 680 tokens/s on a single GB300 in agent workloads. Weights are Apache 2.0 on Hugging Face, ModelScope and GitHub, with SGLang and llama.cpp support.
Key Takeaways
- A 35B-total, roughly 3B-active MoE diffusion model initialized from Qwen3.6-35B-A3B, with 128K context and a thinking mode
- DiffuSpace says it tops all six agent benchmarks it compared: SWE-bench Verified 73.6, Terminal-Bench 2.1 51.7 and tau3 Overall 65.3, ahead of LLaDA2.2 flash, Mercury 2.5 and Qwen3.6-35B-A3B
- Speed (vendor-reported): 680 tokens/s on one GB300 at BF16 and batch size 1 in agent workloads, and 515 tokens/s on one H200 versus 365 tokens/s for Qwen with DFlash2
- An INT4 build runs on a 16GB RTX 5070 Ti at 101.8 tokens/s on MATH-500
- Weights on Hugging Face are Apache 2.0, about 69GB in BF16, and SGLang is the recommended serving path
Linked Models & Agents:
⚡ Clover-35B-A3B-Preview⚡ Qwen3.6-35B-A3B⚡ Dream 7B⚡ LLaDA2.2 flash⚡ Mercury 2.5🌐 DiffuSpace🌐 HKU NLP🌐 Alibaba Qwen🌐 NVIDIA
