InterleaveThinker: Reinforcing Agentic Interleaved Generation
Open paper
Benchmark
Model
Rank
Results
image-editing-on-risebench
InterleaveThinker + Qwen-Image-Edit-2511
#3
RISEBench Overall Accuracy: 30.0
image-generation-on-wise
InterleaveThinker + FLUX.2-klein-9B
#11
Overall: 0.73
Rank counts only results with a code link.