FriendliAI Tech & Research Blog
Supercharge Generative AI Serving: cut costs with the fastest serving engine on the market
Accelerating LLM Training with Memory-Balanced Pipeline Parallelism
Accelerating LLM Training with Memory-Balanced Pipeline Parallelism
PeriFlow’s Enriched Coverage for Sought-After LLMs: MPT, LLaMA, and Dolly
PeriFlow’s Enriched Coverage for Sought-After LLMs: MPT, LLaMA, and Dolly
We have some exciting news to share!
Get an Extra Speedup of LLM Inference with Integer Quantization on PeriFlow
Get an Extra Speedup of LLM Inference with Integer Quantization on PeriFlow
At FriendliAI, our top priority is to deliver a serving system with the best performance. We are excited to introduce a new feature that…
Fine-tuning and Serving CodeGen, a Code Generation Model, with PeriFlow
Fine-tuning and Serving CodeGen, a Code Generation Model, with PeriFlow
CodeGen, unveiled in 2022 by Salesforce, is a language model that allows users to create programs with natural language instead of having…
Save on Training Costs of Generative AI with PeriFlow
Save on Training Costs of Generative AI with PeriFlow
Generative AI is already widely used for chatbots, translation, code generation, summarization, image generation, and much more. Thanks to…
Serve generative AI models like T5 faster than ever with PeriFlow (32.8x faster for T5–3B)
Serve generative AI models like T5 faster than ever with PeriFlow (32.8x faster for T5–3B)
In our previous blog articles (#1, #2), we showed the performance gain of PeriFlow (aka Orca) on GPT3, a popular generative AI model. Orca…