ray-project / llm-numbers

Numbers every LLM developer should know

ray-project/llm-numbers Issues

Typical energy consumption / emitted CO2 of an LLM for serving
Updated 2 months ago1
Be Concise - Lower Quality?
Updated 2 months ago
Cost Efficiency
Closed 3 months ago
CPU Stats for when it's possible
Updated 5 months ago2
Typical run times of an LLM for serving
Updated 5 months ago
Understanding 1MB per token calculation
Updated 6 months ago1
Cost optimization with "Be Concise" instruction is not clear.
Updated 9 months ago
Calculation of "5:1 -- Cost Ratio of generation of text using GPT-3.5-Turbo vs OpenAI embedding"
Updated 9 months ago3
llm-numbers-zh_CN（中文版）
Updated 9 months ago
Users want to see history / trend of LLM numbers
Updated a year ago
Put two numbers in your ratio
Closed a year ago2
Inconsistent Order of Cost Ratio in README
Closed a year ago2
accuracy?
Closed a year ago1
50
Closed a year ago3
Idea: Track these numbers over time
Closed a year ago1
Token to word ratio is incorrect in cheat sheet
Closed a year ago2
Fine tune cost numbers
Closed a year ago2