wasmtime

Library Listed

Wasmtime implementation of the wasi-nn API

Checked ยท Sources

Downloads
22,358
Dependents
6
Stars
18,835
Forks
1,920
Contributors
413
Commits
17,696
Latest release
50.0.1 ยท

About

From the project README ยท

A standalone runtime for WebAssembly

A Bytecode Alliance project

Guide | Contributing | Website | Chat

Installation

The Wasmtime CLI can be installed on Linux and macOS (locally) with a small install script:

curl https://wasmtime.dev/install.sh -sSf | bash

This script installs into $WASMTIME_HOME (defaults to $HOME/.wasmtime), and executable is placed in $WASMTIME_HOME/bin.

After running the install script above, follow the on-screen instructions.

Windows or otherwise interested users can download installers and binaries directly from the GitHub Releases page.

For additional installation options, refer to the online book CLI installation page.

Documentation on Wasmtime's currently supported versions can be found in the online book documentation.

Example

If you've got the Rust compiler installed then you can take some Rust source code:

fn main() {
    println!("Hello, world!");
}

and compile it into a WebAssembly component with:

rustup target add wasm32-wasip2
rustc hello.rs --target wasm32-wasip2

Once compiled, you can run your component:

wasmtime hello.wasm

You should see the following output:

Hello, world!
Read the full README

(Note: make sure you installed Rust using the rustup method in the official instructions above, and do not have a copy of the Rust toolchain installed on your system in some other way as well (e.g. the system package manager). Otherwise, the rustup target add... command may not install the target for the correct copy of Rust.)

Features

  • Fast. Wasmtime is built on the optimizing Cranelift code generator to quickly generate high-quality machine code either at runtime or ahead-of-time. Wasmtime is optimized for efficient instantiation, low-overhead calls between the embedder and wasm, and scalability of concurrent instances.

  • Secure. Wasmtime's development is strongly focused on correctness and security. Building on top of Rust's runtime safety guarantees, each Wasmtime feature goes through careful review and consideration via an RFC process. Once features are designed and implemented, they undergo 24/7 fuzzing donated by Google's OSS Fuzz. As features stabilize they become part of a release, and when things go wrong we have a well-defined security policy in place to quickly mitigate and patch any issues. We follow best practices for defense-in-depth and integrate protections and mitigations for issues like Spectre. Finally, we're working to push the state-of-the-art by collaborating with academic researchers to formally verify critical parts of Wasmtime and Cranelift.

  • Configurable. Wasmtime uses sensible defaults, but can also be configured to provide more fine-grained control over things like CPU and memory consumption. Whether you want to run Wasmtime in a tiny environment or on massive servers with many concurrent instances, we've got you covered.

  • WASI. Wasmtime supports a rich set of APIs for interacting with the host environment through the WASI standard.

  • Standards Compliant. Wasmtime passes the official WebAssembly test suite, implements the official C API of wasm, and implements future proposals to WebAssembly as well. Wasmtime developers are intimately engaged with the WebAssembly standards process all along the way too.

Language Support

You can use Wasmtime from a variety of different languages through embeddings of the implementation.

Languages supported by the Bytecode Alliance:

Languages supported by the community:

Documentation

๐Ÿ“š Read the Wasmtime guide here! ๐Ÿ“š

The wasmtime guide is the best starting point to learn about what Wasmtime can do for you or help answer your questions about Wasmtime. If you're curious in contributing to Wasmtime, it can also help you do that!


It's Wasmtime.

Runs on
CUDA
What shows it
  • CUDA: wasmtime-wasi-nn feature pytorch (dependency tch)
Native code
Optional
  • Binds native code: openvino-sys (feature openvino), ort-sys (feature onnx), torch-sys (feature pytorch)
Python bindings
No Python binding crate found
License
Apache-2.0 WITH LLVM-exception

Compare with similar projects

Order: strongest tier first, then most downloads, then name, with unmeasured values last. A library counts downloads over 30 days. An app counts binary downloads per release. An app with none counts its crate downloads over 30 days. ML Rust runs this catalogue, so its closest project always gets a row.

wasmtime beside the projects that share a topic with it
ProjectDownloadsDependentsStarsLatest releaseRuns on (GPU)Native codeLicense
wasmtime This page Listed

Wasmtime implementation of the wasi-nn API

22,358 downloads over 30 days 6 18,835 50.0.1 CUDA Binds native code: openvino-sys (feature openvino), ort-sys (feature onnx), torch-sys (feature pytorch) Apache-2.0 WITH LLVM-exception
rten Featured

Tensor library for the RTen machine learning runtime

496,924 downloads over 30 days 29 360 0.27.0 GPU support unknown No native code found Apache-2.0 OR MIT
llama-gguf Listed

A high-performance Rust implementation of llama.cpp - LLM inference engine with full GGUF support

1,787 downloads over 30 days 3 21 0.14.0 CUDA, Metal, Vulkan Binds native code: hailort-sys (feature hailo), libsqlite3-sys (feature rag-sqlite) Apache-2.0 OR MIT
mesh-llm App Listed

Distributed AI/LLM for the people. Share compute privately or publicly to power your agents and chat.

1,411 binary downloads per release Not applicable 3,993 0.78.1 GPU support unknown Binds native code: libsqlite3-sys, netlink-sys, ring, secp256k1-sys Apache-2.0 OR MIT
LlamaEdge App Listed

The easiest & fastest way to run customized and fine-tuned LLMs locally or on the edge

841 binary downloads per release Not applicable 1,663 Not measured Not measured Apache-2.0
blazr App Listed

Inference server for LLMs, speech and embeddings: OpenAI-compatible routes with streaming, plus speech synthesis (TTS) and transcription (ASR). Serves Llama, Mistral, Qwen, Mamba2, MLA+MoE and hybrid models from SafeTensors, AWQ, GPTQ and GGUF.

8 binary downloads per release Not applicable 7 0.7.1 CUDA, WebGPU Binds native code: ring (feature distributed) Apache-2.0

600 KB WASM runtime for Cactus Compute's Needle AI tool-calling models, Needle 3, 2 and 1 from one build. Browser, Cloudflare Workers, Node.js, Python, C FFI and no_std. Token-exact with the JAX reference. No backend, no API key.

126 downloads over 30 days 0 107 0.3.1 GPU support unknown No native code found MIT
Why ListedNo archive, New or Featured rule places it, so it is Listed.

What it lacks for Featured

  • It has 3 active dependents and 2 new dependents in 365 days. Featured needs 11 active dependents or 9 new dependents in 365 days. The repository route does not pass: the repo's main crate wasmtime is not in this entry.

What else the check found

  • The last activity is on 7 Nov 2026, 2 days ago. The rules expect an update within 365 days. Dormancy rule

Exact rules: Listed, Featured library. The check on applied thresholds version 9.

Every rule outcome, as the check records it
  • expected-interval: version 50.0.1 base 300, raised to the 365-day floor: E = 365 days
  • dormancy: last activity 2026-11-07, 2 days ago, E = 365: active
  • featured: not featured: active_users 3 < 11 and new_adopters 2 < 9 and repository route: the repo's main crate wasmtime is not in this entry
  • listed: listed: no archive, new or featured rule applies
Crates2 crates

Measurements and sources

Each signal row shows the raw value, its source and the fetch date. The Sources list gathers every distinct source on this page. Movement needs more history.

Sources6 sources
Adoption3 active dependents, 2 new dependents in 365 days1 not measured
Adoption measurements
SignalValueSourceFetchedCounts
Active dependents3crates.io database dumpYes
New dependents (365 days)2crates.io database dumpYes
Peak active dependents3crates.io database dumpYes
Active dependents as a share of the peak1crates.io database dumpYes
Binary downloads per releaseNot measuredYes
MaintenanceLatest human commit: 2 days ago3 not measured
Maintenance measurements
SignalValueSourceFetchedCounts
Latest human commit2026-11-07GitHub repositoryYes
Latest release2026-11-05crates.io database dumpYes
First release2021-01-08crates.io database dumpYes
Latest version50.0.1crates.io database dumpYes
Days the oldest outside issue waits for a reply17GitHub repositoryYes
Maintainer replies over 365 days29GitHub repositoryYes
Repository archivedfalseGitHub repositoryYes
Repository gonefalseGitHub repositoryYes
Release commit or manifest in the repotruecrates.io packageYes
Published docstruecrates.io database dumpYes
Latest release fails cargo checkNot measuredYes
Build check dateNot measuredYes
Build check toolchainNot measuredYes
SecurityUnpatched vulnerabilities: none
Security measurements
SignalValueSourceFetchedCounts
Unpatched vulnerabilities past the grace windownoneRustSec advisory databaseYes
RustSec unmaintained advisorynoneRustSec advisory databaseYes
Deprecation noticenonecrates.io database dumpYes
Deprecated requirementnonecrates.io database dumpYes
Distribution223 releases, 22,358 downloads over 30 days5 not measured
Distribution measurements
SignalValueSourceFetchedCounts
Releases223crates.io database dumpYes
Downloads over 30 days22358crates.io database dumpYes
Binary downloads over the last 3 releasesNot measuredYes
Homebrew installs on request over 365 daysNot measuredYes
Repology repository familiesNot measuredYes
Docker pull growth over 90 daysNot measuredYes
Docker pulls at the growth baselineNot measuredYes
Every version a prereleasefalsecrates.io database dumpYes
Every version 0.0.xfalsecrates.io database dumpYes
Every version yankedfalsecrates.io database dumpYes
Crate deleted from crates.iofalsecrates.io database dumpYes
Repository18,835 stars, 1,920 forks, 413 contributors, 17,696 commits. Stars alone never qualify a project.
Repository measurements
SignalValueSourceFetchedCounts
GitHub stars18835GitHub repositoryYes โ€” one part of the repository route
GitHub forks1920GitHub repositoryYes โ€” one part of the repository route
Contributors with a GitHub account413GitHub contributorsYes โ€” one part of the repository route
Commits on the default branch17696GitHub repositoryNo
Repository's main crate left out of the entrywasmtimecrates.io database dumpYes
Facts that never count
Facts that never count measurements
SignalValueSourceFetchedCounts
LicenseApache-2.0 WITH LLVM-exceptioncrates.io database dumpNo
Runs on, from listed crates and their direct dependenciesCUDAcrates.io database dumpNo
What shows each backendCUDA: wasmtime-wasi-nn feature pytorch (dependency tch)crates.io database dumpNo
Native code, from listed crates and their dependenciesBinds native code: openvino-sys (feature openvino), ort-sys (feature onnx), torch-sys (feature pytorch)crates.io database dumpNo
Python bindings, from direct dependencies of listed cratesNo Python binding crate foundcrates.io database dumpNo
Crates that depend on a listed crate directly6crates.io database dumpNo

Corrections and review history

Report a correction

No correction applies to this entry.

Judgments

A judgment enters only through an approved issue and a merged pull request.

No judgment applies to this entry.