rwkv-architecture
RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.
Details
- Path
- cli-tool/components/skills/ai-research/model-architecture-rwkv/SKILL.md
- License
- MIT
- Dependencies
- 4