cactus
Run AI locally on phones, wearables and AI-native hardware
Stars: 2973
Cactus is an energy-efficient and fast AI inference framework designed for phones, wearables, and resource-constrained arm-based devices. It provides a bottom-up approach with no dependencies, optimizing for budget and mid-range phones. The framework includes Cactus FFI for integration, Cactus Engine for high-level transformer inference, Cactus Graph for unified computation graph, and Cactus Kernels for low-level ARM-specific operations. It is suitable for implementing custom models and scientific computing on mobile devices.
README:
Energy-efficient & fast AI inference framework for phones, wearables and AI-native hardware. Budget and mid-range phones control over 70% of the market, but frameworks today optimise for the highend phones with advanced chips. Cactus is designed bottom-up with no dependencies, for resource-constrained arm-based devices.
Example (CPU-only):
- Model: Qwen3-600m-INT8
- File size: 370-420mb
- 16-20 t/s on Pixel 6a, Galaxy S21, iPhone 11 Pro
- 50-70 t/s on Pixel 9, Galaxy S25, iPhone 16
Cactus exposes 4 levels of abstraction.
┌─────────────────┐
│ Cactus FFI │ ←── OpenAI compatible C API for integration
└─────────────────┘
│
┌─────────────────┐
│ Cactus Engine │ ←── High-level transformer engine
└─────────────────┘
│
┌─────────────────┐
│ Cactus Graph │ ←── Unified zero-copy computation graph
└─────────────────┘
│
┌─────────────────┐
│ Cactus Kernels │ ←── Low-level ARM-specific SIMD operations
└─────────────────┘
Cactus Graph is a general numerical computing framework that runs on Cactus Kernels. Great for implementing custom models and scientific computing, like JAX for phones.
#include cactus.h
CactusGraph graph;
auto a = graph.input({2, 3}, Precision::FP16);
auto b = graph.input({3, 4}, Precision::INT8);
auto x1 = graph.matmul(a, b, false);
auto x2 = graph.transpose(x1);
auto result = graph.matmul(b, x2, true);
float a_data[6] = {1.1f, 2.3f, 3.4f, 4.2f, 5.7f, 6.8f};
float b_data[12] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12};
graph.set_input(a, a_data, Precision::FP16);
graph.set_input(b, b_data, Precision::INT8);
graph.execute();
void* output_data = graph.get_output(result);
graph.hard_reset();
Cactus Engine is an transformer inference engine built on top of Cactus Graphs. It is abstracted via the minimalist Cactus Foreign Function Interface.
#include cactus.h
const char* model_path = "path/to/weight/folder";
cactus_model_t model = cactus_init(model_path, 2048);
const char* messages = R"([
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "/nothink My name is Henry Ndubuaku"}
])";
const char* options = R"({
"temperature": 0.1,
"top_p": 0.95,
"top_k": 20,
"max_tokens": 50,
"stop_sequences": ["<|im_end|>"]
})";
char response[1024];
int result = cactus_complete(model, messages, response, sizeof(response), options, nullptr, nullptr);This makes it easy to write Cactus bindings for any language. Header files are self-documenting for each but documentation contributions are welcome.
Cactus SDKs run 500k+ weekly inference tasks in production today, try them!
You can run these codes directly on Macbooks with Apple chips due to their design. Performance gain is observed in mobile devices but for testing during development, Vanilla M3 CPU-only can run Qwen3 600m INT8 at ~55 toks/sec, use the following:
- Generate weights from HuggingFace model:
python3 tools/convert_hf.py Qwen/Qwen3-0.6B weights/qwen3-600m-i8/ --precision INT8- Build and test:
./tests/run.sh # remember to chmod +x any script first time
- Gemma, SmolVLM, Liquid, Kitten, Vosk etc.
- SMMLA, NPU & DSP for high-end phones.
- INT4 support for 1B+ models.
- Python tools for porting Torch/JAX cactus.
Preliminary results:
- Qwen3-4B-INT4 on iPhone 16 Pro NPU = 21 t/s
While Cactus can be used for all Apple devices including Macbooks, for computers/AMD/Intel/Nvidia generally, please use HuggingFace, Llama.cpp, Ollama, vLLM, MLX. They're built for those, support x86, and are all great!
For Tasks:
Click tags to check more tools for each tasksFor Jobs:
Alternative AI tools for cactus
Similar Open Source Tools
cactus
Cactus is an energy-efficient and fast AI inference framework designed for phones, wearables, and resource-constrained arm-based devices. It provides a bottom-up approach with no dependencies, optimizing for budget and mid-range phones. The framework includes Cactus FFI for integration, Cactus Engine for high-level transformer inference, Cactus Graph for unified computation graph, and Cactus Kernels for low-level ARM-specific operations. It is suitable for implementing custom models and scientific computing on mobile devices.
lancedb
LanceDB is an open-source database for vector-search built with persistent storage, which greatly simplifies retrieval, filtering, and management of embeddings. The key features of LanceDB include: Production-scale vector search with no servers to manage. Store, query, and filter vectors, metadata, and multi-modal data (text, images, videos, point clouds, and more). Support for vector similarity search, full-text search, and SQL. Native Python and Javascript/Typescript support. Zero-copy, automatic versioning, manage versions of your data without needing extra infrastructure. GPU support in building vector index(*). Ecosystem integrations with LangChain 🦜️🔗, LlamaIndex 🦙, Apache-Arrow, Pandas, Polars, DuckDB, and more on the way. LanceDB's core is written in Rust 🦀 and is built using Lance, an open-source columnar format designed for performant ML workloads.
trpc-agent-go
A powerful Go framework for building intelligent agent systems with large language models (LLMs), hierarchical planners, memory, telemetry, and a rich tool ecosystem. tRPC-Agent-Go enables the creation of autonomous or semi-autonomous agents that reason, call tools, collaborate with sub-agents, and maintain long-term state. The framework provides detailed documentation, examples, and tools for accelerating the development of AI applications.
catalyst
Catalyst is a C# Natural Language Processing library designed for speed, inspired by spaCy's design. It provides pre-trained models, support for training word and document embeddings, and flexible entity recognition models. The library is fast, modern, and pure-C#, supporting .NET standard 2.0. It is cross-platform, running on Windows, Linux, macOS, and ARM. Catalyst offers non-destructive tokenization, named entity recognition, part-of-speech tagging, language detection, and efficient binary serialization. It includes pre-built models for language packages and lemmatization. Users can store and load models using streams. Getting started with Catalyst involves installing its NuGet Package and setting the storage to use the online repository. The library supports lazy loading of models from disk or online. Users can take advantage of C# lazy evaluation and native multi-threading support to process documents in parallel. Training a new FastText word2vec embedding model is straightforward, and Catalyst also provides algorithms for fast embedding search and dimensionality reduction.
KULLM
KULLM (구름) is a Korean Large Language Model developed by Korea University NLP & AI Lab and HIAI Research Institute. It is based on the upstage/SOLAR-10.7B-v1.0 model and has been fine-tuned for instruction. The model has been trained on 8×A100 GPUs and is capable of generating responses in Korean language. KULLM exhibits hallucination and repetition phenomena due to its decoding strategy. Users should be cautious as the model may produce inaccurate or harmful results. Performance may vary in benchmarks without a fixed system prompt.
stable-diffusion.cpp
The stable-diffusion.cpp repository provides an implementation for inferring stable diffusion in pure C/C++. It offers features such as support for different versions of stable diffusion, lightweight and dependency-free implementation, various quantization support, memory-efficient CPU inference, GPU acceleration, and more. Users can download the built executable program or build it manually. The repository also includes instructions for downloading weights, building from scratch, using different acceleration methods, running the tool, converting weights, and utilizing various features like Flash Attention, ESRGAN upscaling, PhotoMaker support, and more. Additionally, it mentions future TODOs and provides information on memory requirements, bindings, UIs, contributors, and references.
wandb
Weights & Biases (W&B) is a platform that helps users build better machine learning models faster by tracking and visualizing all components of the machine learning pipeline, from datasets to production models. It offers tools for tracking, debugging, evaluating, and monitoring machine learning applications. W&B provides integrations with popular frameworks like PyTorch, TensorFlow/Keras, Hugging Face Transformers, PyTorch Lightning, XGBoost, and Sci-Kit Learn. Users can easily log metrics, visualize performance, and compare experiments using W&B. The platform also supports hosting options in the cloud or on private infrastructure, making it versatile for various deployment needs.
zeta
Zeta is a tool designed to build state-of-the-art AI models faster by providing modular, high-performance, and scalable building blocks. It addresses the common issues faced while working with neural nets, such as chaotic codebases, lack of modularity, and low performance modules. Zeta emphasizes usability, modularity, and performance, and is currently used in hundreds of models across various GitHub repositories. It enables users to prototype, train, optimize, and deploy the latest SOTA neural nets into production. The tool offers various modules like FlashAttention, SwiGLUStacked, RelativePositionBias, FeedForward, BitLinear, PalmE, Unet, VisionEmbeddings, niva, FusedDenseGELUDense, FusedDropoutLayerNorm, MambaBlock, Film, hyper_optimize, DPO, and ZetaCloud for different tasks in AI model development.
beyondllm
Beyond LLM offers an all-in-one toolkit for experimentation, evaluation, and deployment of Retrieval-Augmented Generation (RAG) systems. It simplifies the process with automated integration, customizable evaluation metrics, and support for various Large Language Models (LLMs) tailored to specific needs. The aim is to reduce LLM hallucination risks and enhance reliability.
VinAI_Translate
VinAI_Translate is a Vietnamese-English Neural Machine Translation System offering state-of-the-art text-to-text translation models for Vietnamese-to-English and English-to-Vietnamese. The system includes pre-trained models with different configurations and parameters, allowing for further fine-tuning. Users can interact with the models through the VinAI Translate system website or the HuggingFace space 'VinAI Translate'. Evaluation scripts are available for assessing the translation quality. The tool can be used in the 'transformers' library for Vietnamese-to-English and English-to-Vietnamese translations, supporting both GPU-based batch translation and CPU-based sequence translation examples.
Eco2AI
Eco2AI is a python library for CO2 emission tracking that monitors energy consumption of CPU & GPU devices and estimates equivalent carbon emissions based on regional emission coefficients. Users can easily integrate Eco2AI into their Python scripts by adding a few lines of code. The library records emissions data and device information in a local file, providing detailed session logs with project names, experiment descriptions, start times, durations, power consumption, CO2 emissions, CPU and GPU names, operating systems, and countries.
AgentFly
AgentFly is an extensible framework for building LLM agents with reinforcement learning. It supports multi-turn training by adapting traditional RL methods with token-level masking. It features a decorator-based interface for defining tools and reward functions, enabling seamless extension and ease of use. To support high-throughput training, it implemented asynchronous execution of tool calls and reward computations, and designed a centralized resource management system for scalable environment coordination. A suite of prebuilt tools and environments are provided.
ShannonBase
ShannonBase is a HTAP database provided by Shannon Data AI, designed for big data and AI. It extends MySQL with native embedding support, machine learning capabilities, a JavaScript engine, and a columnar storage engine. ShannonBase supports multimodal data types and natively integrates LightGBM for training and prediction. It leverages embedding algorithms and vector data type for ML/RAG tasks, providing Zero Data Movement, Native Performance Optimization, and Seamless SQL Integration. The tool includes a lightweight JavaScript engine for writing stored procedures in SQL or JavaScript.
MarkLLM
MarkLLM is an open-source toolkit designed for watermarking technologies within large language models (LLMs). It simplifies access, understanding, and assessment of watermarking technologies, supporting various algorithms, visualization tools, and evaluation modules. The toolkit aids researchers and the community in ensuring the authenticity and origin of machine-generated text.
TornadoVM
TornadoVM is a plug-in to OpenJDK and GraalVM that allows programmers to automatically run Java programs on heterogeneous hardware. TornadoVM targets OpenCL, PTX and SPIR-V compatible devices which include multi-core CPUs, dedicated GPUs (Intel, NVIDIA, AMD), integrated GPUs (Intel HD Graphics and ARM Mali), and FPGAs (Intel and Xilinx).
ichigo
Ichigo is a local real-time voice AI tool that uses an early fusion technique to extend a text-based LLM to have native 'listening' ability. It is an open research experiment with improved multiturn capabilities and the ability to refuse processing inaudible queries. The tool is designed for open data, open weight, on-device Siri-like functionality, inspired by Meta's Chameleon paper. Ichigo offers a web UI demo and Gradio web UI for users to interact with the tool. It has achieved enhanced MMLU scores, stronger context handling, advanced noise management, and improved multi-turn capabilities for a robust user experience.
For similar tasks
cactus
Cactus is an energy-efficient and fast AI inference framework designed for phones, wearables, and resource-constrained arm-based devices. It provides a bottom-up approach with no dependencies, optimizing for budget and mid-range phones. The framework includes Cactus FFI for integration, Cactus Engine for high-level transformer inference, Cactus Graph for unified computation graph, and Cactus Kernels for low-level ARM-specific operations. It is suitable for implementing custom models and scientific computing on mobile devices.
CopilotKit
CopilotKit is an open-source framework for building, deploying, and operating fully custom AI Copilots, including in-app AI chatbots, AI agents, and AI Textareas. It provides a set of components and entry points that allow developers to easily integrate AI capabilities into their applications. CopilotKit is designed to be flexible and extensible, so developers can tailor it to their specific needs. It supports a variety of use cases, including providing app-aware AI chatbots that can interact with the application state and take action, drop-in replacements for textareas with AI-assisted text generation, and in-app agents that can access real-time application context and take action within the application.
kitchenai
KitchenAI is an open-source toolkit designed to simplify AI development by serving as an AI backend and LLMOps solution. It aims to empower developers to focus on delivering results without being bogged down by AI infrastructure complexities. With features like simplifying AI integration, providing an AI backend, and empowering developers, KitchenAI streamlines the process of turning AI experiments into production-ready APIs. It offers built-in LLMOps features, is framework-agnostic and extensible, and enables faster time-to-production. KitchenAI is suitable for application developers, AI developers & data scientists, and platform & infra engineers, allowing them to seamlessly integrate AI into apps, deploy custom AI techniques, and optimize AI services with a modular framework. The toolkit eliminates the need to build APIs and infrastructure from scratch, making it easier to deploy AI code as production-ready APIs in minutes. KitchenAI also provides observability, tracing, and evaluation tools, and offers a Docker-first deployment approach for scalability and confidence.
react-native-executorch
React Native ExecuTorch is a framework that allows developers to run AI models on mobile devices using React Native. It bridges the gap between React Native and native platform capabilities, providing high-performance AI model execution without requiring deep knowledge of native code or machine learning internals. The tool supports ready-made models in `.pte` format and offers a Python API for custom models. It is designed to simplify the integration of AI features into React Native apps.
Mortal
Mortal (凡夫) is a free and open source AI for Japanese mahjong, powered by deep reinforcement learning. It provides a comprehensive solution for playing Japanese mahjong with AI assistance. The project focuses on utilizing deep reinforcement learning techniques to enhance gameplay and decision-making in Japanese mahjong. Mortal offers a user-friendly interface and detailed documentation to assist users in understanding and utilizing the AI effectively. The project is actively maintained and welcomes contributions from the community to further improve the AI's capabilities and performance.
AIInfra
AIInfra is an open-source project focused on AI infrastructure, specifically targeting large models in distributed clusters, distributed architecture, distributed training, and algorithms related to large models. The project aims to explore and study system design in artificial intelligence and deep learning, with a focus on the hardware and software stack for building AI large model systems. It provides a comprehensive curriculum covering key topics such as system overview, AI computing clusters, communication and storage, cluster containers and cloud-native technologies, distributed training, distributed inference, large model algorithms and data, and applications of large models.
crabml
Crabml is a llama.cpp compatible AI inference engine written in Rust, designed for efficient inference on various platforms with WebGPU support. It focuses on running inference tasks with SIMD acceleration and minimal memory requirements, supporting multiple models and quantization methods. The project is hackable, embeddable, and aims to provide high-performance AI inference capabilities.
For similar jobs
sweep
Sweep is an AI junior developer that turns bugs and feature requests into code changes. It automatically handles developer experience improvements like adding type hints and improving test coverage.
teams-ai
The Teams AI Library is a software development kit (SDK) that helps developers create bots that can interact with Teams and Microsoft 365 applications. It is built on top of the Bot Framework SDK and simplifies the process of developing bots that interact with Teams' artificial intelligence capabilities. The SDK is available for JavaScript/TypeScript, .NET, and Python.
ai-guide
This guide is dedicated to Large Language Models (LLMs) that you can run on your home computer. It assumes your PC is a lower-end, non-gaming setup.
classifai
Supercharge WordPress Content Workflows and Engagement with Artificial Intelligence. Tap into leading cloud-based services like OpenAI, Microsoft Azure AI, Google Gemini and IBM Watson to augment your WordPress-powered websites. Publish content faster while improving SEO performance and increasing audience engagement. ClassifAI integrates Artificial Intelligence and Machine Learning technologies to lighten your workload and eliminate tedious tasks, giving you more time to create original content that matters.
chatbot-ui
Chatbot UI is an open-source AI chat app that allows users to create and deploy their own AI chatbots. It is easy to use and can be customized to fit any need. Chatbot UI is perfect for businesses, developers, and anyone who wants to create a chatbot.
BricksLLM
BricksLLM is a cloud native AI gateway written in Go. Currently, it provides native support for OpenAI, Anthropic, Azure OpenAI and vLLM. BricksLLM aims to provide enterprise level infrastructure that can power any LLM production use cases. Here are some use cases for BricksLLM: * Set LLM usage limits for users on different pricing tiers * Track LLM usage on a per user and per organization basis * Block or redact requests containing PIIs * Improve LLM reliability with failovers, retries and caching * Distribute API keys with rate limits and cost limits for internal development/production use cases * Distribute API keys with rate limits and cost limits for students
uAgents
uAgents is a Python library developed by Fetch.ai that allows for the creation of autonomous AI agents. These agents can perform various tasks on a schedule or take action on various events. uAgents are easy to create and manage, and they are connected to a fast-growing network of other uAgents. They are also secure, with cryptographically secured messages and wallets.
griptape
Griptape is a modular Python framework for building AI-powered applications that securely connect to your enterprise data and APIs. It offers developers the ability to maintain control and flexibility at every step. Griptape's core components include Structures (Agents, Pipelines, and Workflows), Tasks, Tools, Memory (Conversation Memory, Task Memory, and Meta Memory), Drivers (Prompt and Embedding Drivers, Vector Store Drivers, Image Generation Drivers, Image Query Drivers, SQL Drivers, Web Scraper Drivers, and Conversation Memory Drivers), Engines (Query Engines, Extraction Engines, Summary Engines, Image Generation Engines, and Image Query Engines), and additional components (Rulesets, Loaders, Artifacts, Chunkers, and Tokenizers). Griptape enables developers to create AI-powered applications with ease and efficiency.