Today for AI
HOT RADAR
llmHEAT 11.3°

AI CURATED HOT STORY · 10/28/2026

Clover-35B Open-Sourced: Diffusion LLM Uses MoE to Challenge Agentic Coding

1 reports archived1 independent sourcesupdated 10/28/2026, 23:16:10
Synthesis & Latest Updates
1 Sources Cross-Validated

Shenzhen-based DiffuSpace has released and open-sourced Clover-35B-A3B-Preview, the first large-scale diffusion language model (dLLM) to integrate a Mixture-of-Experts (MoE) architecture. With 35B total parameters and 3B active parameters, it achieves a generation speed of 680 tokens/s, aiming to break autoregressive serial bottlenecks via parallel decoding while demonstrating competitive or superior performance in certain agentic tasks compared to same-size AR models.

LATEST/Shenzhen-based DiffuSpace has released and open-sourced Clover-35B-A3B-Preview, the first large-scale diffusion language model (dLLM) to integrate a Mixture-of-Experts (MoE) architecture. With 35B total parameters and 3B active parameters, it achieves a generation speed of 680 tokens/s, aiming to break autoregressive serial bottlenecks via parallel decoding while demonstrating competitive or superior performance in certain agentic tasks compared to same-size AR models.

Hourly heat curve

2 fully observed hours
Fewer than 3 fully observed hours; trend not plotted yet.

Coverage timeline

Total 1 reports · Latest first
  1. InfoQ 中文T2·85 pts

    Clover-35B Open-Sourced: Diffusion LLM Uses MoE to Challenge Agentic Coding

    • First introduction of MoE architecture into large-scale diffusion language models, balancing parameter scale and inference efficiency
    • Generation speed reaches 680 tokens/s, significantly outperforming traditional autoregressive serial decoding
    • Performance meets or exceeds same-size autoregressive models in certain agentic tasks and general benchmarks