RWKV-5 1.5B
by RWKV Foundation · open-source · Last verified 2026-03-17
RWKV-5 (Eagle) is a 1.5 billion parameter model from the RWKV architecture family that combines the parallelism of transformers during training with the efficiency of RNNs during inference. It achieves linear time and constant memory complexity, making it exceptionally efficient for long-context tasks and edge deployment without the quadratic cost of attention mechanisms.
https://huggingface.co/RWKV/v5-Eagle-World-1B5-v2-20240520-ctx4096 ↗C
C—Below Average
Adoption: CQuality: BFreshness: BCitations: C+Engagement: F
Specifications
- License
- Apache 2.0
- Pricing
- open-source
- Capabilities
- text-generation, long-context-processing, constant-memory-inference, multilingual-generation
- Integrations
- Hugging Face, RWKV.cpp, Ollama
- Use Cases
- long document processing, streaming inference on edge, multilingual assistants, memory-constrained deployment
- API Available
- Yes
- Parameters
- ~1.5B
- Context Window
- 4K
- Modalities
- text
- Training Cutoff
- 2024
- Tags
- rwkv, rnn, efficient, open-source, linear-attention
- Added
- 2026-03-17
- Completeness
- 100%
Index Score
44.7Adoption
42
Quality
67
Freshness
68
Citations
58
Engagement
0