ModelLLMsv2.0

Eagle 2

by RWKV Foundation · free · Last verified 2026-03-17

Eagle 2 is the second-generation RWKV Eagle model, a linear-complexity recurrent neural network architecture that achieves competitive Transformer-level performance on language benchmarks while maintaining constant inference memory usage. It enables efficient deployment on edge devices and resource-constrained environments without sacrificing quality.

https://huggingface.co/RWKV/v6-Finch-7B-HF ↗

D—Poor

Adoption: DQuality: B+Freshness: ACitations: DEngagement: F

Specifications

License: Apache 2.0
Pricing: free
Capabilities: text-generation, efficient-inference, long-context, instruction-following
Integrations: Hugging Face, RWKV.cpp, Ollama
Use Cases: edge-deployment, low-latency-inference, long-context-processing, research
API Available: Yes
Parameters: 7B
Context Window: Infinite (recurrent)
Modalities: text
Training Cutoff: 2024
Tags: rwkv, linear-rnn, open-source, efficient, recurrent
Added: 2026-03-17
Completeness: 100%

Index Score

36.4

Adoption

Quality

Freshness

Citations

Engagement

Explore the full AI ecosystem on Agents as a Service