
LLMcalc
A tool to determine whether or not your PC can run a given LLM
Stars: 122

LLM Calculator is a script that estimates the memory requirements and performance of Hugging Face models based on quantization levels. It fetches model parameters, calculates required memory, and analyzes performance with different RAM/VRAM configurations. The tool supports Windows and Linux, AMD, Intel, and Nvidia GPUs. Users can input a Hugging Face model ID to get its parameter count and analyze memory requirements for various quantization schemes. The tool provides estimates for GPU offload percentage and throughput in tk/s. It requires dependencies like python, uv, pciutils for AMD + Linux, and drivers for Nvidia. The tool is designed for rough estimates and may not work with MultiGPU setups.
README:
This script estimates the memory requirements and performance of Hugging Face models based on quantization levels. It fetches model parameters, calculates required memory, and analyzes performance with different RAM/VRAM configurations.
It supports windows and Linux, AMD, Intel, and Nvidia GPUs. You will need smi (cuda toolkit (?)) installed to recognise your Nvidia GPU.
Warning: The tool isn't tested outside of Linux+Nvidia, so results may be inaccurate. It's a rough estimate. It may or may not work with MultiGPU setups. If not, use -n to specify how many cards you have (assuming they're all the same). If you have an RTX 3070 and an RTX 3060 for example, you may need to specify -v and -b to be the average values of the two.
Flags
-b, --bandwidth: Override memory bandwidth (GB/s).
-n, --num-gpus: Number of GPUs (default is 1).
-v, --vram: Override VRAM amount per card in GB.
Example:
python main.py -b 950 -n 2 -v 24
You'll need these to run it in python. 3.12.3 is what I wrote it in, but any modern version should work.
Install uv and run with:
uv run main.py
For AMD + Linux you'll need sudo apt install pciutils
Nvidia will need drivers, as long as nvidia-smi
works this program should.
Intel needs lspci
, dunno if that supports windows.
Enter a Hugging Face model ID (e.g., microsoft/phi-4) to get its parameter count. The script fetches system RAM and VRAM specs. You can override them with flags. It analyzes memory requirements for several quantization schemes and estimates throughput (tk/s).
Hover over a cell to see how many layers you need to offload, and what context you can fit without KV cache quantizing.
Cli Output
Enter Hugging Face model ID (e.g., microsoft/phi-4): microsoft/phi-4
Model Parameters: 14.7B params (14.70B params)
Total RAM: 33.53 GB
VRAM: 8.00 GB, ~448.0GB/s
Estimated RAM Bandwidth: 64.00 GB/s
Analysis for each quantization level:
FP8:
Run Type: Partial offload
Memory Required: 16.43 GB
GPU Offload Percentage: 48.7%
Estimated tk/s: 5.38
Q6_K_S:
Run Type: Partial offload
Memory Required: 13.86 GB
GPU Offload Percentage: 57.7%
Estimated tk/s: 7.39
Q5_K_S:
Run Type: Partial offload
Memory Required: 11.84 GB
GPU Offload Percentage: 67.6%
Estimated tk/s: 10.63
Q4_K_M:
Run Type: Partial offload
Memory Required: 10.55 GB
GPU Offload Percentage: 75.8%
Estimated tk/s: 14.71
IQ4_XS:
Run Type: Partial offload
Memory Required: 9.64 GB
GPU Offload Percentage: 83.0%
Estimated tk/s: 19.92
Q3_K_M:
Run Type: KV cache offload
Memory Required: 8.90 GB
Estimated tk/s: 45.30
IQ3_XS:
Run Type: All in VRAM
Memory Required: 7.80 GB
Estimated tk/s: 57.45
IQ2_XS:
Run Type: All in VRAM
Memory Required: 6.14 GB
Estimated tk/s: 72.90
For Tasks:
Click tags to check more tools for each tasksFor Jobs:
Alternative AI tools for LLMcalc
Similar Open Source Tools

LLMcalc
LLM Calculator is a script that estimates the memory requirements and performance of Hugging Face models based on quantization levels. It fetches model parameters, calculates required memory, and analyzes performance with different RAM/VRAM configurations. The tool supports Windows and Linux, AMD, Intel, and Nvidia GPUs. Users can input a Hugging Face model ID to get its parameter count and analyze memory requirements for various quantization schemes. The tool provides estimates for GPU offload percentage and throughput in tk/s. It requires dependencies like python, uv, pciutils for AMD + Linux, and drivers for Nvidia. The tool is designed for rough estimates and may not work with MultiGPU setups.

mac-studio-server
This repository provides configuration and scripts for running Ollama LLM server on Apple Silicon Macs in headless mode, optimized for performance and resource usage. It includes features like automatic startup, system resource optimization, external network access, proper logging setup, and SSH-based remote management. Users can customize the Ollama service configuration and enable optional GPU memory optimization and Docker autostart for container applications. The installation process disables unnecessary system services, configures power management, and optimizes for background operation while maintaining Screen Sharing capability for remote management. Performance considerations focus on reducing memory usage, disabling GUI-related services, minimizing background processes, preventing sleep/hibernation, and optimizing for headless operation.

Easy-Translate
Easy-Translate is a script designed for translating large text files with a single command. It supports various models like M2M100, NLLB200, SeamlessM4T, LLaMA, and Bloom. The tool is beginner-friendly and offers seamless and customizable features for advanced users. It allows acceleration on CPU, multi-CPU, GPU, multi-GPU, and TPU, with support for different precisions and decoding strategies. Easy-Translate also provides an evaluation script for translations. Built on HuggingFace's Transformers and Accelerate library, it supports prompt usage and loading huge models efficiently.

python-aiplatform
The Vertex AI SDK for Python is a library that provides a convenient way to use the Vertex AI API. It offers a high-level interface for creating and managing Vertex AI resources, such as datasets, models, and endpoints. The SDK also provides support for training and deploying custom models, as well as using AutoML models. With the Vertex AI SDK for Python, you can quickly and easily build and deploy machine learning models on Vertex AI.

gpt4all
GPT4All is an ecosystem to run powerful and customized large language models that work locally on consumer grade CPUs and any GPU. Note that your CPU needs to support AVX or AVX2 instructions. Learn more in the documentation. A GPT4All model is a 3GB - 8GB file that you can download and plug into the GPT4All open-source ecosystem software. Nomic AI supports and maintains this software ecosystem to enforce quality and security alongside spearheading the effort to allow any person or enterprise to easily train and deploy their own on-edge large language models.

snd
Sales & Dungeons is a tool that utilizes thermal printers for creating customizable handouts, quick references, and more for Dungeons and Dragons sessions. It offers extensive templating and random generation systems, supports various connection methods, and allows importing/exporting templates and data sources. Users can access external data sources like Open5e, import data from CSV and other formats, and utilize AI prompt generation and translation. The tool supports cloud sync and is compatible with multiple operating systems and devices.

agents
Polymarket Agents is a developer framework and set of utilities for building AI agents to trade autonomously on Polymarket. It integrates with Polymarket API, provides AI agent utilities for prediction markets, supports local and remote RAG, sources data from various services, and offers comprehensive LLM tools for prompt engineering. The architecture features modular components like APIs and scripts for managing local environments, server set-up, and CLI for end-user commands.

zipnn
ZipNN is a lossless and near-lossless compression library optimized for numbers/tensors in the Foundation Models environment. It automatically prepares data for compression based on its type, allowing users to focus on core tasks without worrying about compression complexities. The library delivers effective compression techniques for different data types and structures, achieving high compression ratios and rates. ZipNN supports various compression methods like ZSTD, lz4, and snappy, and provides ready-made scripts for file compression/decompression. Users can also manually import the package to compress and decompress data. The library offers advanced configuration options for customization and validation tests for different input and compression types.

BentoML
BentoML is an open-source model serving library for building performant and scalable AI applications with Python. It comes with everything you need for serving optimization, model packaging, and production deployment.

djl
Deep Java Library (DJL) is an open-source, high-level, engine-agnostic Java framework for deep learning. It is designed to be easy to get started with and simple to use for Java developers. DJL provides a native Java development experience and allows users to integrate machine learning and deep learning models with their Java applications. The framework is deep learning engine agnostic, enabling users to switch engines at any point for optimal performance. DJL's ergonomic API interface guides users with best practices to accomplish deep learning tasks, such as running inference and training neural networks.

labo
LABO is a time series forecasting and analysis framework that integrates pre-trained and fine-tuned LLMs with multi-domain agent-based systems. It allows users to create and tune agents easily for various scenarios, such as stock market trend prediction and web public opinion analysis. LABO requires a specific runtime environment setup, including system requirements, Python environment, dependency installations, and configurations. Users can fine-tune their own models using LABO's Low-Rank Adaptation (LoRA) for computational efficiency and continuous model updates. Additionally, LABO provides a Python library for building model training pipelines and customizing agents for specific tasks.

skyrim
Skyrim is a weather forecasting tool that enables users to run large weather models using consumer-grade GPUs. It provides access to state-of-the-art foundational weather models through a well-maintained infrastructure. Users can forecast weather conditions, such as wind speed and direction, by running simulations on their own GPUs or using modal volume or cloud services like s3 buckets. Skyrim supports various large weather models like Graphcast, Pangu, Fourcastnet, and DLWP, with plans for future enhancements like ensemble prediction and model quantization.

agentok
Agentok Studio is a tool built upon AG2, a powerful agent framework from Microsoft, offering intuitive visual tools to streamline the creation and management of complex agent-based workflows. It simplifies the process for creators and developers by generating native Python code with minimal dependencies, enabling users to create self-contained code that can be executed anywhere. The tool is currently under development and not recommended for production use, but contributions are welcome from the community to enhance its capabilities and functionalities.

OpenAdapt
OpenAdapt is an open-source software adapter between Large Multimodal Models (LMMs) and traditional desktop and web Graphical User Interfaces (GUIs). It aims to automate repetitive GUI workflows by leveraging the power of LMMs. OpenAdapt records user input and screenshots, converts them into tokenized format, and generates synthetic input via transformer model completions. It also analyzes recordings to generate task trees and replay synthetic input to complete tasks. OpenAdapt is model agnostic and generates prompts automatically by learning from human demonstration, ensuring that agents are grounded in existing processes and mitigating hallucinations. It works with all types of desktop GUIs, including virtualized and web, and is open source under the MIT license.

open-parse
Open Parse is a Python library for visually discerning document layouts and chunking them effectively. It is designed to fill the gap in open-source libraries for handling complex documents. Unlike text splitting, which converts a file to raw text and slices it up, Open Parse visually analyzes documents for superior LLM input. It also supports basic markdown for parsing headings, bold, and italics, and has high-precision table support, extracting tables into clean Markdown formats with accuracy that surpasses traditional tools. Open Parse is extensible, allowing users to easily implement their own post-processing steps. It is also intuitive, with great editor support and completion everywhere, making it easy to use and learn.

scalene
Scalene is a high-performance CPU, GPU, and memory profiler for Python that provides detailed information and runs faster than many other profilers. It incorporates AI-powered proposed optimizations, allowing users to generate optimization suggestions by clicking on specific lines or regions of code. Scalene separates time spent in Python from native code, highlights hotspots, and identifies memory usage per line. It supports GPU profiling on NVIDIA-based systems and detects memory leaks. Users can generate reduced profiles, profile specific functions using decorators, and suspend/resume profiling for background processes. Scalene is available as a pip or conda package and works on various platforms. It offers features like profiling at the line level, memory trends, copy volume reporting, and leak detection.
For similar tasks

Efficient-Multimodal-LLMs-Survey
Efficient Multimodal Large Language Models: A Survey provides a comprehensive review of efficient and lightweight Multimodal Large Language Models (MLLMs), focusing on model size reduction and cost efficiency for edge computing scenarios. The survey covers the timeline of efficient MLLMs, research on efficient structures and strategies, and applications. It discusses current limitations and future directions in efficient MLLM research.

uvadlc_notebooks
The UvA Deep Learning Tutorials repository contains a series of Jupyter notebooks designed to help understand theoretical concepts from lectures by providing corresponding implementations. The notebooks cover topics such as optimization techniques, transformers, graph neural networks, and more. They aim to teach details of the PyTorch framework, including PyTorch Lightning, with alternative translations to JAX+Flax. The tutorials are integrated as official tutorials of PyTorch Lightning and are relevant for graded assignments and exams.

LiveBench
LiveBench is a benchmark tool designed for Language Model Models (LLMs) with a focus on limiting contamination through monthly new questions based on recent datasets, arXiv papers, news articles, and IMDb movie synopses. It provides verifiable, objective ground-truth answers for accurate scoring without an LLM judge. The tool offers 18 diverse tasks across 6 categories and promises to release more challenging tasks over time. LiveBench is built on FastChat's llm_judge module and incorporates code from LiveCodeBench and IFEval.

farel-bench
The 'farel-bench' project is a benchmark tool for testing LLM reasoning abilities with family relationship quizzes. It generates quizzes based on family relationships of varying degrees and measures the accuracy of large language models in solving these quizzes. The project provides scripts for generating quizzes, running models locally or via APIs, and calculating benchmark metrics. The quizzes are designed to test logical reasoning skills using family relationship concepts, with the goal of evaluating the performance of language models in this specific domain.

LLMcalc
LLM Calculator is a script that estimates the memory requirements and performance of Hugging Face models based on quantization levels. It fetches model parameters, calculates required memory, and analyzes performance with different RAM/VRAM configurations. The tool supports Windows and Linux, AMD, Intel, and Nvidia GPUs. Users can input a Hugging Face model ID to get its parameter count and analyze memory requirements for various quantization schemes. The tool provides estimates for GPU offload percentage and throughput in tk/s. It requires dependencies like python, uv, pciutils for AMD + Linux, and drivers for Nvidia. The tool is designed for rough estimates and may not work with MultiGPU setups.
For similar jobs

weave
Weave is a toolkit for developing Generative AI applications, built by Weights & Biases. With Weave, you can log and debug language model inputs, outputs, and traces; build rigorous, apples-to-apples evaluations for language model use cases; and organize all the information generated across the LLM workflow, from experimentation to evaluations to production. Weave aims to bring rigor, best-practices, and composability to the inherently experimental process of developing Generative AI software, without introducing cognitive overhead.

LLMStack
LLMStack is a no-code platform for building generative AI agents, workflows, and chatbots. It allows users to connect their own data, internal tools, and GPT-powered models without any coding experience. LLMStack can be deployed to the cloud or on-premise and can be accessed via HTTP API or triggered from Slack or Discord.

VisionCraft
The VisionCraft API is a free API for using over 100 different AI models. From images to sound.

kaito
Kaito is an operator that automates the AI/ML inference model deployment in a Kubernetes cluster. It manages large model files using container images, avoids tuning deployment parameters to fit GPU hardware by providing preset configurations, auto-provisions GPU nodes based on model requirements, and hosts large model images in the public Microsoft Container Registry (MCR) if the license allows. Using Kaito, the workflow of onboarding large AI inference models in Kubernetes is largely simplified.

PyRIT
PyRIT is an open access automation framework designed to empower security professionals and ML engineers to red team foundation models and their applications. It automates AI Red Teaming tasks to allow operators to focus on more complicated and time-consuming tasks and can also identify security harms such as misuse (e.g., malware generation, jailbreaking), and privacy harms (e.g., identity theft). The goal is to allow researchers to have a baseline of how well their model and entire inference pipeline is doing against different harm categories and to be able to compare that baseline to future iterations of their model. This allows them to have empirical data on how well their model is doing today, and detect any degradation of performance based on future improvements.

tabby
Tabby is a self-hosted AI coding assistant, offering an open-source and on-premises alternative to GitHub Copilot. It boasts several key features: * Self-contained, with no need for a DBMS or cloud service. * OpenAPI interface, easy to integrate with existing infrastructure (e.g Cloud IDE). * Supports consumer-grade GPUs.

spear
SPEAR (Simulator for Photorealistic Embodied AI Research) is a powerful tool for training embodied agents. It features 300 unique virtual indoor environments with 2,566 unique rooms and 17,234 unique objects that can be manipulated individually. Each environment is designed by a professional artist and features detailed geometry, photorealistic materials, and a unique floor plan and object layout. SPEAR is implemented as Unreal Engine assets and provides an OpenAI Gym interface for interacting with the environments via Python.

Magick
Magick is a groundbreaking visual AIDE (Artificial Intelligence Development Environment) for no-code data pipelines and multimodal agents. Magick can connect to other services and comes with nodes and templates well-suited for intelligent agents, chatbots, complex reasoning systems and realistic characters.