LLaDA2.2-flash

LLaDA2.2-flash is an agent-oriented Mixture-of-Experts (MoE) diffusion language model designed for long-context agentic workloads. It introduces Levenshtein Editing with DELETE and INSERT control tokens to enable efficient parallel generation, error correction, and multi-turn tool use.

LLaDA2.2-flash is an agent-oriented Mixture-of-Experts (MoE) diffusion language model designed for long-context agentic workloads. It introduces Levenshtein Editing with DELETE and INSERT control tokens to enable efficient parallel generation, error correction, and multi-turn tool use.

• 01128K context window with Block Routing for efficient MoE expert activation
• 02Levenshtein Editing using DELETE and INSERT control tokens for structural sequence modification
• 03Agentic Reinforcement Learning via L-EBPO for improved error correction
• 04100B parameter Mixture-of-Experts (MoE) architecture
• 05High-throughput parallel diffusion decoding