GitHub · Python
caseclose/cma-harness
Cognitive-structured Multimodal Agent (CMA-Harness): a memory-centric agent for long-horizon multimodal understanding, generation, and editing — externalizing visual history into episodic memory with selective retrieval. Includes the M2CA-Bench benchmark.
- Stars
- 16
- Forks
- 2
- Stars gained in 24 hours
- 0
- Forks gained in 24 hours
- 0
Topics: agent, benchmark, image-editing, image-generation, llm, memory, multimodal, retrieval
Latest data point: Nov 21, 2026 · 16 stars · 2 forks