SDSAT: Accelerating LLM Inference through Speculative Decoding with Semantic Adaptive Tokens
Paper
•
2403.18647
•
Published
Model Card for Model ID
The 13B model of "SDSAT: Accelerating LLM Inference through Speculative Decoding with Semantic Adaptive Tokens"
Model Details
Model Description
- Developed by: ainergy
- Language(s) (NLP): Code
- Finetuned from model: CodeLlama-13B
Model Sources
- Repository: https://github.com/ainergy-ml/SDSAT
- Paper: https://arxiv.org/abs/2403.18647
Evaluation
Results
Walltime improvement
- Downloads last month
- 2