RahulSChand/gpu_poor
Web-based calculator that estimates GPU memory requirements, token/s throughput, and fine-tuning time for any LLM.
Not currently ranked — collecting fresh signals.
star history
A JavaScript tool that helps users determine if their GPU can run a specific LLM and what performance to expect. It calculates vRAM usage broken down by model size, KV cache, activation memory, and quantization overhead for GGML, bitsandbytes, and QLoRA methods. It also estimates token generation speed and fine-tuning iteration times, supporting inference frameworks like vLLM, llama.cpp, and HuggingFace.
Frequently asked
- What is RahulSChand/gpu_poor?
- Web-based calculator that estimates GPU memory requirements, token/s throughput, and fine-tuning time for any LLM.
- Is gpu_poor open source?
- Yes — RahulSChand/gpu_poor is an open-source project tracked on heatdrop.
- What language is gpu_poor written in?
- RahulSChand/gpu_poor is primarily written in JavaScript.
- How popular is gpu_poor?
- RahulSChand/gpu_poor has 1.4k stars on GitHub.
- Where can I find gpu_poor?
- RahulSChand/gpu_poor is on GitHub at https://github.com/RahulSChand/gpu_poor.