Search Results for "metal"
Sort By:
Showing 25 open source projects for "metal"
View related business solutions-
$300 Free Credits to Build on Google CloudPut your $300 in credit toward real workloads, then keep building with free monthly usage for 20+ products. No commitment and no charge until you upgrade.
-
Build Data Resilience - Take the Assessment TodayIs your recovery strategy as strong as you think? Take this quick self-assessment to check your recovery readiness and gain tailored insights. In only 2 minutes, you'll learn where you fall on the recovery readiness scale.
-
1
h3-metal
MiniMax H3 inference engine for Mac computers
h3-metal is a native MiniMax-H3 inference engine for Apple Silicon focused on generating video and audio locally with Metal. It supports prompt-to-video and prompt-to-audio workflows as well as first-frame, last-frame, image, video, and audio references. An interactive terminal session keeps prompt conditioning, the diffusion transformer, and the video decoder in memory for faster repeated generations. -
2
tt-metal
TT-NN operator library, and TT-Metalium low level kernel programming
tt-metal, also referred to in its documentation as TT-Metalium, is Tenstorrent’s low-level software development kit for programming applications on Tenstorrent AI accelerators. The project is designed for developers who need direct access to the company’s Tensix processor architecture, exposing a programming model that is closer to hardware control than high-level inference frameworks. -
3
Beta9
Run serverless GPU workloads with fast cold starts on bare-metal
beta9 is a platform that enables running serverless GPU workloads with fast cold starts on bare-metal servers globally. It allows developers to deploy and scale GPU-accelerated applications without managing underlying infrastructure, offering flexibility and efficiency for AI and high-performance computing tasks. beta9 supports various frameworks and provides tools for monitoring and managing deployments effectively. -
4
ds4.c
DeepSeek 4 Flash local inference engine for Metal
ds4.c is a specialized local inference engine created by antirez for running DeepSeek V4 Flash models directly on Apple Silicon hardware using Metal acceleration. Unlike general-purpose inference runtimes, the project is intentionally optimized for a specific model family, enabling highly efficient execution and simplified architecture. The engine includes DS4-specific model loading, KV cache management, prompt rendering, and OpenAI-compatible server APIs for local deployment workflows. ... -
Save Up to 91% on Cloud Compute With Spot VMsRun batch jobs at 60-91% off with Spot VMs. Long-running workloads get automatic discounts with sustained use.
-
5
TurboFieldfare
Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook
TurboFieldfare is a custom Swift and Metal runtime for running the instruction-tuned Gemma 4 26B-A4B model on Apple Silicon Macs with limited memory. Instead of loading the entire model, it keeps the shared core and KV cache in RAM while streaming only the routed experts required for each token from SSD. This approach reduces active memory use to roughly 2 GB while the installed model occupies about 14.3 GB of storage. -
6
stable-diffusion.cpp
Diffusion model(SD,Flux,Wan,Qwen Image,Z-Image,...) inference
...It enables text-to-image and image-to-image generation, supports a growing set of models like SD1.x, SD2.x, SDXL, SD-Turbo, Qwen Image, and more, and is continually updated with support for cutting-edge model variants including video and image editing models. The project is built on the ggml backend, which allows efficient execution on CPUs and GPUs via backends like CUDA, Vulkan, Metal, OpenCL, and SYCL, making it suitable for everything from desktops to mobile devices. It includes options for ControlNet, LoRA models, upscaling via ESRGAN, and advanced sampling techniques, giving developers and users a rich toolkit for creative workflows. -
7
OpenVINO Model Server
A scalable inference server for models optimized with OpenVINO
...The server exposes model inference via standard network protocols like REST and gRPC, allowing any client that speaks those protocols to request predictions remotely, abstracting away the complexity of where and how the model runs. It supports model deployment in diverse environments including Docker, bare-metal machines, and Kubernetes clusters, and is especially useful in microservices architectures where AI services need to scale independently. The system supports a wide range of model sources, letting you host models from local storage, remote object storage, or even pull from model hubs. -
8
BaseRT
Fastest LLM inference runtime for Apple Silicon
BaseRT is a local large language model inference runtime optimized for Apple Silicon computers. It accelerates model execution through hand-written Metal kernels and requires an M1 or newer Mac running macOS 14 or later. A unified command-line interface can download models from Hugging Face, convert checkpoints, launch chats, benchmark performance, and inspect model packages. Its server implements OpenAI-compatible chat, completion, embedding, transcription, tool-calling, and multimodal endpoints. ... -
9
gpt-oss
gpt-oss-120b and gpt-oss-20b are two open-weight language models
...Both models use a native MXFP4 quantization for efficient memory use and support OpenAI’s Harmony response format, enabling transparent full chain-of-thought reasoning and advanced tool integrations such as function calling, browsing, and Python code execution. The repository provides multiple reference implementations—including PyTorch, Triton, and Metal—for educational and experimental use, as well as example clients and tools like a terminal chat app and a Responses API server. -
MongoDB Atlas runs apps anywhereMongoDB Atlas gives you the freedom to build and run modern applications anywhere—across AWS, Azure, and Google Cloud. With global availability in over 115 regions, Atlas lets you deploy close to your users, meet compliance needs, and scale with confidence across any geography.
-
10
llama.cpp
LLM inference in C/C++
...It provides command-line tools, a server mode with an OpenAI-compatible API style, model conversion utilities, and extensive backend acceleration options. llama.cpp runs on CPUs and GPUs, with support for Apple silicon, x86, RISC-V, CUDA, HIP, Vulkan, SYCL, Metal, and hybrid CPU-GPU execution. Its main value is making practical LLM inference accessible across consumer machines, servers, and specialized deployment environments. -
11
Bonsai 27B
Run Bonsai (1-bit) and Ternary-Bonsai language models locally
...It supports the 1-bit Bonsai and higher-quality Ternary-Bonsai families in 1.7B, 4B, 8B, and 27B sizes. The models can run on macOS, Linux, and Windows through CPU, Metal, CUDA, Vulkan, ROCm, llama.cpp, or MLX backends. Its 27B models process text, images, screenshots, and PDFs while supporting reasoning and long-context conversations. They also provide OpenAI-compatible tool calling and optional MCP server integration for agentic workflows. Automated setup scripts install dependencies, download models, obtain binaries, and configure interactive interfaces. ... -
12
qvac-fabric-llm.cpp
QVAC Fabric: cross-platform LLM inference and fine-tuning
...The project focuses on removing hardware limitations traditionally associated with LLM deployment by enabling support for a wide range of backends, including Vulkan, Metal, CUDA, and CPU, making it accessible on devices ranging from smartphones to enterprise servers. It introduces native LoRA fine-tuning capabilities that can be executed directly on consumer hardware, allowing developers to train and adapt models locally without relying on cloud infrastructure. A key innovation is its support for BitNet ternary quantized models, enabling highly efficient inference and training even on resource-constrained systems. -
13
Nexa SDK
Nexa SDK is a comprehensive toolkit for supporting ONNX and GGML
...Additionally, it offers an OpenAI-compatible API server with JSON schema mode for function calling and streaming support, and a user-friendly Streamlit UI. Users can run Nexa SDK in any device with Python environment, and GPU acceleration is supported, including CUDA, Metal, and ROCm. An executable version is also available. -
14
OuteTTS
Interface for OuteTTS models
...The project supports multiple backends including llama.cpp (Python bindings and server), Hugging Face Transformers, ExLlamaV2, VLLM and a JavaScript interface via Transformers.js, allowing it to run on CPUs, NVIDIA CUDA GPUs, AMD ROCm, Vulkan-capable GPUs, and Apple Metal. It also includes a notion of speaker profiles: you can create a speaker from a short audio sample, save it as JSON, and reuse it for consistent voice identity across generations and sessions. For best quality, the model is designed to work with a reference speaker clip and will inherit emotion, style, and accent from that reference. -
15
MLC LLM
Universal LLM Deployment Engine with ML Compilation
...The system supports deployment on environments including Linux, macOS, Windows, iOS, Android, and web browsers while utilizing different acceleration technologies such as CUDA, Vulkan, Metal, and WebGPU. It also provides OpenAI-compatible APIs that allow developers to integrate locally deployed models into existing AI applications without major code changes. -
16
MimiClaw
Run OpenClaw on a $5 chip
MimiClaw (from the mimiclaw project) is an edge-AI personal assistant that runs directly on extremely low-cost hardware like an ESP32-S3 microcontroller without a full operating system, Node.js, or cloud backend. By running pure C on a bare-metal chip, MimiClaw brings AI interactions and persistent memory to a tiny USB-powered device you can carry in your pocket. You connect the device to Wi-Fi and chat with it using Telegram, making it a convenient always-on assistant for tasks like reminders, quick lookups, or custom AI interactions. Even though it’s running on minimal hardware, MimiClaw maintains local memory that persists across power cycles, enabling context continuity over time without relying on cloud services. ... -
17
ANE Training
Training neural networks on Apple Neural Engine via APIs
...The repository implements a from-scratch transformer training pipeline capable of running both forward and backward passes on ANE hardware without relying on CoreML, Metal, or GPU acceleration. It explores the internal software stack of the Apple Neural Engine by interfacing with private classes such as _ANEClient and compiling custom compute graphs in the MIL format. The project includes performance benchmarks and kernel breakdowns that show how different components of the training loop are distributed between the ANE and CPU. ... -
18
MNN
MNN is a blazing fast, lightweight deep learning framework
MNN is a highly efficient and lightweight deep learning framework. It supports inference and training of deep learning models, and has industry leading performance for inference and training on-device. At present, MNN has been integrated in more than 20 apps of Alibaba Inc, such as Taobao, Tmall, Youku, Dingtalk, Xianyu and etc., covering more than 70 usage scenarios such as live broadcast, short video capture, search recommendation, product searching by image, interactive marketing, equity... -
19
llm.c
LLM training in simple, raw C/CUDA
...Portability is a goal: it aims to compile with common toolchains and run on modest hardware for small experiments. Rather than delivering a production-grade stack, it serves as a reference and learning scaffold for people who want to “see the metal” behind LLMs. -
20
uzu
A high-performance inference engine for AI models
uzu is a high-performance inference engine designed to run artificial intelligence models efficiently on Apple Silicon hardware. Written primarily in Rust and leveraging Apple’s Metal framework, the project focuses on maximizing performance when executing large language models and other AI workloads on devices such as Mac computers with M-series chips. The engine implements a hybrid architecture in which model layers can be executed either as custom GPU kernels or through Apple’s MPSGraph API, allowing it to balance performance and compatibility depending on the workload. ... -
21
Elkeid
Open source solution that can meet the requirements of workloads
Elkeid is an open-source platform for security and intrusion-detection that aims to support a wide variety of deployment contexts — from bare-metal hosts to containers, Kubernetes clusters, and even serverless environments. It was born out of ByteDance’s internal security best practices, offering for community users a subset of its enterprise-grade capabilities. Elkeid combines kernel-level data collection, user-space agents, and runtime instrumentation (RASP) to detect malicious behavior, file anomalies, runtime exploits, and suspicious container activity. ... -
22
Dead Deer 3.14.92.2026
3D modeler, 3D game maker, 3D demo maker
...Android .NED Player (install APK and "open with" with file managers) APK generator for Android. Support for: Direct3D9 (SM3) Direct3D10 (SM4) Direct3D11 (SM5) Direct3D12 (SM5) OpenGL and GLSL OpenGLES 2/3 Apple METAL Retina, UHD. Intel x86/64, ARMv7/ARM64, RISCV. Linux (Ubuntu/wxWidgets(Gtk3)). iOS /iPasOS (with XCode) (GLES20/METAL) Windows Phone Windows VR (Steam/Oculus) WebAsm/WebGL UWP Windows/XBOX SDL2 Linux ARM 32/64 RISCV OpenXR (Quest?/Pico) 3.14.92.2026 -
23
FastChat
Open platform for training, serving, and evaluating language models
...If you do not have enough memory, you can enable 8-bit compression by adding --load-8bit to the commands above. This can reduce memory usage by around half with slightly degraded model quality. It is compatible with the CPU, GPU, and Metal backend. Vicuna-13B with 8-bit compression can run on a single NVIDIA 3090/4080/T4/V100(16GB) GPU. In addition to that, you can add --cpu-offloading to commands above to offload weights that don't fit on your GPU onto the CPU memory. This requires 8-bit compression to be enabled and the bitsandbytes package to be installed, which is only available on linux operating systems. -
24
llm
An ecosystem of Rust libraries for working with large language models
llm is an ecosystem of Rust libraries for working with large language models - it's built on top of the fast, efficient GGML library for machine learning. The primary entry point for developers is the llm crate, which wraps the llm-base and the supported model crates. Documentation for the released version is available on Docs.rs. For end-users, there is a CLI application, llm-cli, which provides a convenient interface for interacting with supported models. Text generation can be done as a... -
25
The CDM-Core ontology (CREMA Data Model - core module) is a manufacturing ontology, specialised in the two application sub-domains of 1) exhaust car manufacturing and 2) metallic press maintenance The CDM-Core ontology was developed by: Dr. Luca Mazzola, Msc. Patrick Kapahnke, Marko Vujic, and PD Dr. Matthias Klusch at the German Research Center for Artificial Intelligence DFKI GmbH (http://www.dfki.de ) in Saarbrücken, Germany. Copyright: DFKI, 2016, licence CC BY-SA see...
- Previous
- You're on page 1
- Next