Pre-release

This release is a pre-release and may not be stable for production use.

A Native Implementation of Telegram Calls in a seamless way.
Examples • Documentation • PyPi • npm • crates.io • Maven Central • Releases • Channel • Chat

NTgCalls

NTgCalls is a lightweight open-source library for media streaming in Telegram calls. Built from scratch in C++ with WebRTC & Boost, it prioritizes accessibility to developers and resource efficiency.

Powerful	Simple	Light
Built from scratch in C++ using Boost and libwebrtc	Simple Python, Node.js, Rust, GO, C Bindings and Java for Android SDK	We removed anything that could burden the library, keeping every binding a thin layer over the same native core

Build Status

Architecture	Windows	Linux	MacOS
x86_64
ARM64

Features

Pre-built binaries for macOS (arm64-v8a), Linux (x86_64, arm64-v8a), Windows (x86_64), and Android (x86, 86_64, arm64-v8a, armv7)
Call flexibility: Group and private call support
Media controls: pause/resume and mute/unmute
Codec compatibility: H.264, HEVC (H.265), VP8, VP9, AV1, AAC, MP3, Opus
Content sharing: Screen streaming, Microphone and Camera streaming
Pre-built wheels for Python, npm packages for Node.js, crates for Rust & AAR SDK library for Android

Installing

Prebuilt packages are published for every supported language, so most users never need to compile anything.

Language	Package	Install
Python	ntgcalls	`pip install ntgcalls`
Node.js	ntgcalls	`npm install ntgcalls`
Rust	ntgcalls	`cargo add ntgcalls`
Java (Android)	io.github.pytgcalls:ntgcalls	Gradle dependency
C, C++ and Go	Release archives	Download the shared or static zip

Building from source

Prerequisites are listed in the Build Guide. All commands are run from the root of the repository.

Target	Command	Output
Python	`python3 setup.py install`	Installed in the current environment
Node.js	`python3 setup.py build_lib --target=node`	`targets/node/build/Release/ntgcalls.node`
C and C++	`python3 setup.py build_lib`	`shared-output` and `static-output`

The C target builds both the shared and the static library in a single run. Each output directory also contains an include folder with the headers to add to your project.

The Rust bindings do not build the native library: the ntgcalls-sys build script downloads the matching static libraries from the GitHub release for your platform. Set NTGCALLS_RELEASE_BASE to build against a local release instead.

Using the library from Go

[!WARNING] Static linking for Windows is not supported yet since our library is built with MSVC and Go uses MinGW for static linking. More info can be found here

Go consumes the C bindings through an example project in ./examples/go/:

Download the shared or static release from the Releases page
Copy ntgcalls.h into ./examples/go/ntgcalls/
Copy the remaining files into ./examples/go/
- ntgcalls.dll or ntgcalls.lib on Windows amd64
- libntgcalls.so or libntgcalls.a on Linux amd64
- libntgcalls.dylib or libntgcalls.a on macOS
Run go build or go run . with CGO_ENABLED=1
- $env:CGO_ENABLED=1; go run . on Windows PowerShell
- CGO_ENABLED=1 go run . on UNIX

Key Contributors

@Laky-64 (DevOps Engineer, Software Architect, Porting Engineer):
- Played a crucial role in developing NTgCalls.
- Created the Python Bindings that made the library accessible to Python developers.
- Developed the C Bindings, enabling the library's use in various environments.
@dadadani (Senior C++ Developer, Performance engineer):
- Contributed to setting up CMakeLists and integrating with pybind11, greatly simplifying the library's usage for C++ developers.
@kuogi (Senior UI/UX designer, Documenter):
- As a Senior UI/UX Designer, Kuogi has significantly improved the user interface of our documentation, making it more visually appealing and user-friendly.
- It Has also played a key role in writing and structuring our documentation, ensuring that it is clear, informative, and accessible to all users.
@vrumger (Mid-level NodeJS Developer):
- Avrumy has made important fixes and enhancements to the WebRTC component of the library, improving its stability and performance.
@ankit-chaubey (Rust Developer):
- Authored the first Rust bindings for NTgCalls and generously transferred the ntgcalls crate to the official project.

Junior Developers

@TuriOG (Junior Python Developer):
- Currently working on integrating NTgCalls into PyTgCalls, an important step in expanding the functionality and usability of the library.
@doggyhaha (Junior DevOps, Tester):
- Performs testing of NTgCalls on Linux to ensure its reliability and compatibility.
- Specializes in creating and maintaining GitHub Actions, focusing on automation tasks.
@tappo03 (Junior Go Developer, Tester):
- Performs testing of NTgCalls on Windows to ensure its reliability and compatibility.
- It Is in the process of integrating NTgCalls into a Go wrapper, further enhancing the library's versatility and accessibility.

Special Thanks

@shiguredo: We extend our special thanks to 時雨堂 (shiguredo) for their invaluable assistance in integrating the WebRTC component. Their contributions, using the Sora C++ SDK, have been instrumental in enhancing the functionality of our library.
@evgeny-nadymov: A heartfelt thank you to Evgeny Nadymov for graciously allowing us to use their code from telegram-react. His contribution has been pivotal to the success of this project.
@morethanwords: We extend our special thanks to morethanwords for their invaluable help in integrating the connection to WebRTC with Telegram Web K. Their assistance has been instrumental in enhancing the functionality of our library.
@MarshalX: for their generous assistance in answering questions and providing insights regarding WebRTC.
@LyzCoote: for providing an ARM64 Server and allowing us to build an image with clang-18 preinstalled on manylinux2014 arm64.

We would like to extend a special thanks to @null-nick and @branchscope for their valuable contributions to the testing phase of the library and to @SadLuffy for his assistance as a copywriter. Their dedication to testing and optimizing the library has been instrumental in its success.

Additionally, we extend our gratitude to all contributors for their exceptional work in making this project a reality.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ntgcalls-3.0.0rc1.tar.gz (20.6 MB view details)

Uploaded Sep 2, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

ntgcalls-3.0.0rc1-cp314-cp314t-manylinux_2_28_x86_64.whl (42.8 MB view details)

Uploaded Sep 2, 2026 CPython 3.14tmanylinux: glibc 2.28+ x86-64

ntgcalls-3.0.0rc1-cp314-cp314t-manylinux_2_28_aarch64.whl (38.7 MB view details)

Uploaded Sep 2, 2026 CPython 3.14tmanylinux: glibc 2.28+ ARM64

ntgcalls-3.0.0rc1-cp314-cp314t-macosx_14_0_arm64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.14tmacOS 14.0+ ARM64

ntgcalls-3.0.0rc1-cp314-cp314-win_amd64.whl (11.3 MB view details)

Uploaded Sep 2, 2026 CPython 3.14Windows x86-64

ntgcalls-3.0.0rc1-cp314-cp314-manylinux_2_28_x86_64.whl (42.6 MB view details)

Uploaded Sep 2, 2026 CPython 3.14manylinux: glibc 2.28+ x86-64

ntgcalls-3.0.0rc1-cp314-cp314-manylinux_2_28_aarch64.whl (38.5 MB view details)

Uploaded Sep 2, 2026 CPython 3.14manylinux: glibc 2.28+ ARM64

ntgcalls-3.0.0rc1-cp314-cp314-macosx_14_0_arm64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.14macOS 14.0+ ARM64

ntgcalls-3.0.0rc1-cp313-cp313-win_amd64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.13Windows x86-64

ntgcalls-3.0.0rc1-cp313-cp313-manylinux_2_28_x86_64.whl (42.6 MB view details)

Uploaded Sep 2, 2026 CPython 3.13manylinux: glibc 2.28+ x86-64

ntgcalls-3.0.0rc1-cp313-cp313-manylinux_2_28_aarch64.whl (38.5 MB view details)

Uploaded Sep 2, 2026 CPython 3.13manylinux: glibc 2.28+ ARM64

ntgcalls-3.0.0rc1-cp313-cp313-macosx_14_0_arm64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.13macOS 14.0+ ARM64

ntgcalls-3.0.0rc1-cp312-cp312-win_amd64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.12Windows x86-64

ntgcalls-3.0.0rc1-cp312-cp312-manylinux_2_28_x86_64.whl (42.6 MB view details)

Uploaded Sep 2, 2026 CPython 3.12manylinux: glibc 2.28+ x86-64

ntgcalls-3.0.0rc1-cp312-cp312-manylinux_2_28_aarch64.whl (38.5 MB view details)

Uploaded Sep 2, 2026 CPython 3.12manylinux: glibc 2.28+ ARM64

ntgcalls-3.0.0rc1-cp312-cp312-macosx_14_0_arm64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.12macOS 14.0+ ARM64

ntgcalls-3.0.0rc1-cp311-cp311-win_amd64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.11Windows x86-64

ntgcalls-3.0.0rc1-cp311-cp311-manylinux_2_28_x86_64.whl (42.4 MB view details)

Uploaded Sep 2, 2026 CPython 3.11manylinux: glibc 2.28+ x86-64

ntgcalls-3.0.0rc1-cp311-cp311-manylinux_2_28_aarch64.whl (38.4 MB view details)

Uploaded Sep 2, 2026 CPython 3.11manylinux: glibc 2.28+ ARM64

ntgcalls-3.0.0rc1-cp311-cp311-macosx_14_0_arm64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.11macOS 14.0+ ARM64

ntgcalls-3.0.0rc1-cp310-cp310-win_amd64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.10Windows x86-64

ntgcalls-3.0.0rc1-cp310-cp310-manylinux_2_28_x86_64.whl (42.4 MB view details)

Uploaded Sep 2, 2026 CPython 3.10manylinux: glibc 2.28+ x86-64

ntgcalls-3.0.0rc1-cp310-cp310-manylinux_2_28_aarch64.whl (38.4 MB view details)

Uploaded Sep 2, 2026 CPython 3.10manylinux: glibc 2.28+ ARM64

ntgcalls-3.0.0rc1-cp310-cp310-macosx_14_0_arm64.whl (11.0 MB view details)

Uploaded Sep 2, 2026 CPython 3.10macOS 14.0+ ARM64

File details

Details for the file ntgcalls-3.0.0rc1.tar.gz.

File metadata

Download URL: ntgcalls-3.0.0rc1.tar.gz
Upload date: Sep 2, 2026
Size: 20.6 MB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for ntgcalls-3.0.0rc1.tar.gz
Algorithm	Hash digest
SHA256	`767561add1ab77135edc93a0e65c27e28c4bf0f38d032fe8ce0d8f8e50384b50`
MD5	`8b933b778d4b42bc5aa540d94f3125fd`
BLAKE2b-256	`399a17a54a9dec09bb7896fbcea4ddbd15cff38222aee2ffd90afa4990d421d1`

Algorithm	Hash digest
SHA256	`5300520586db45970e17557dcf9f16aeaf79ba7aa29b9cd9908fa335116d4b83`
MD5	`75934e659b1b5f25c04f49ab81dc31ef`
BLAKE2b-256	`6a959668dffe172adcb34d43cfcdc82ff073b752f25dcc3aea6c4922321700ba`

Algorithm	Hash digest
SHA256	`1047393189f38987ab0a5b1eb90e22e2d258a9088a771792a13cb84bd430fd9a`
MD5	`3c0e8e5f91c8fde4fe23d8d988898966`
BLAKE2b-256	`50245627b7d83fbe438d07a7d253af53bbef6af88f56e7bc7b80809763010e44`

Algorithm	Hash digest
SHA256	`8faa549110bff492c84a7e579ff9880736b3cdd7020dc4fd33cd659e9ce8587b`
MD5	`cb4baad1de7c1783a8eda09f8c5de079`
BLAKE2b-256	`a865e545c13a32f9844ae50b2c0aa1263d41bfcd3057b496c2a13c24077cbb38`

Algorithm	Hash digest
SHA256	`8e072a3720c0e69dd8afef91a460d060f0a72ccd2e8214428a10a4ba4071b1ee`
MD5	`632980d509e01fd1519611a2c833a3c0`
BLAKE2b-256	`9505f160df4fc59da4dad6a41ad2339332353d6504c201f6a28f96f2bec213fd`

Algorithm	Hash digest
SHA256	`3fbb80c05002e9d1f077f7b11e7edb1418b630f0ab6cb9195091d11e74768172`
MD5	`517161afe487cb511e01c38ee445161d`
BLAKE2b-256	`6a079be1c635a2f7997eb414985a29954dc2e82cc7de2cd0037ef9dd5fb31d80`

Algorithm	Hash digest
SHA256	`d9a16fb526d095986d6afa03acd89006f27eab51ec83bd752d4b2caf6e409bf7`
MD5	`690fa9135d4b98af9333c63eb8883797`
BLAKE2b-256	`3e5b7b53b724a350e7ce4305c1ee7259582bcf1008ad625d8242aa7ede35bde3`

Algorithm	Hash digest
SHA256	`b7fe4e8e2d8b65b32ea7cfc26990b09116c4c754a61f214753387f41edfeb694`
MD5	`7255783a95924dd7a9f899eac4d46323`
BLAKE2b-256	`a344d932cd7e16c154e74271004e4d6c3636b9ca2362c2e9feeb45aabbc49de7`

Algorithm	Hash digest
SHA256	`72e5dd74896a3c379b572f0db1328cec9fe2cfd6e7e8aff5d392ada122be6bb7`
MD5	`5cc9046a9b3d05c2e22c0bf9d0983aae`
BLAKE2b-256	`4b29def0020e2c3220a0745bab4346a71cacaa48a97063fa019f4ba73c2baf7d`

Algorithm	Hash digest
SHA256	`faf15c34f6cbd2c1c0704b0718b603c6a5e57461eb71677a3daafd8411d7c4d0`
MD5	`06ca2d34729e3d6d8c186ce952767507`
BLAKE2b-256	`a66a1e637e5c81396ba77fa729e5edc7ee23606a7563fc8495d9d49555b0cff2`

Algorithm	Hash digest
SHA256	`681103dbe83e1d0e9c693b312373c97ba061eace56a0c1bb6dad04f232c0528c`
MD5	`9e8a0b00dcfe1ae6940c4111e64b45c5`
BLAKE2b-256	`ad9bc5f480f2192cd152dd944730d0b6e8009c4b75b68ea8c72860339264216e`

Algorithm	Hash digest
SHA256	`f9b2a2ee245cb24c297a586f8cb8165e6fa245d4b3c8a94bd8ba8a581450df02`
MD5	`0cabeb15c999bb2a049eb82b50a08f1c`
BLAKE2b-256	`ed10c495c86d92d7bdc6d8a9ff9522033724a6cdc9781d920cbbf4302ba8c978`

Algorithm	Hash digest
SHA256	`6a244d4aca69c5af1b51b30cbc040f5cc8c707923a4ecdc1e5ff89bbd4355e8e`
MD5	`27d3f5026671a8baf890a5c7456e1606`
BLAKE2b-256	`edc904033a96a2c3b09f730fbe875072f837da3987419110db2405d29e0f1a40`

Algorithm	Hash digest
SHA256	`5abaab8a04e696babc27092655c931c840a7b92e97f2d1aa550520b77eb44ff2`
MD5	`7909a06a62d2bb649aa9bc7f91832723`
BLAKE2b-256	`5a738f683431035c9df65bf6b12fd2961aa993c3d66ba7ac09bbf201418a40eb`

Algorithm	Hash digest
SHA256	`ced62bf0c6c1613d52e713f67dcdd63737216106a58287c8640ec58b0dd849d9`
MD5	`cf6fd4a126ed5227250526da158f5726`
BLAKE2b-256	`9b9ed218a5dad2c6a7121ce541fbbb9d1419fd915dc4da1defdae5cd355fc80b`

Algorithm	Hash digest
SHA256	`15ecbbae39fb1072c2a0e19f565f26c2dcd557d724acbd60227394acd4b5de90`
MD5	`7ebe841829ead41cfc31d515dd5d4a41`
BLAKE2b-256	`6ef179fec22cd8e49a7968c9e2b4f2818c6eaa9ed5739f7c5626eaffdd9ec249`

Algorithm	Hash digest
SHA256	`1438e018873fc6fcc278a11aca4fad130af105d49f228217caf51309793034bb`
MD5	`af59df45371bf6687dcf85b069b14fee`
BLAKE2b-256	`1aa637b10c127bea1be4d9dde8c1b0e5d1cc43fea3605bdf8c2a10e1ea7408be`

Algorithm	Hash digest
SHA256	`1186b6a226a10b6b3f1bde22deafb5464a60663ebafedbbb2f5981e10a53d1f3`
MD5	`b60eba8808383e7bde2a0f7e27f88423`
BLAKE2b-256	`37b09687aa7283153493b44cb9a7b394f5e99c23a0210219a4398e0c6e8a35c1`

Algorithm	Hash digest
SHA256	`6c77c4e8a011f58f2cc8edf7d9962a787f381965d1a4da3b356972b0496170e1`
MD5	`4199f39491dca5f72619f0bbba8f1fb8`
BLAKE2b-256	`2439202247e690455b2e54e3e38c6965a691c9ff30c030f9ac1da0db199cd39c`

Algorithm	Hash digest
SHA256	`7039b40b4e342b69cbf8711b53d65819bd9dfb9749f6f333ac7642effeeeede9`
MD5	`93e8836fb69d1837dc45b991f537cdf9`
BLAKE2b-256	`1424794a72c64412916648cba92afaea9cd4dcf6815513d36dc103f7f8914866`

Algorithm	Hash digest
SHA256	`eaf41bc1060749b834e78e808d827d28d323f5492c0723c0aa2e6220d6eedfc0`
MD5	`23c5e867e0a93b6f39f8420c717d6483`
BLAKE2b-256	`44f487ca17bcbd8b579efe6c6ce486696532708a853268bc211e9f588c7ca694`

ntgcalls 3.0.0rc1

NTgCalls

Build Status

Features

Installing

Building from source

Key Contributors

Junior Developers

Special Thanks

Project links

Key dates

2 maintainers

Credits

GitHub Statistics

Requires

Provides Extra

Tags

Classifiers

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Algorithm	Hash digest
SHA256	`5e798cb969f562738bea61c20f6b225eb51bfeb9fe3b5e517d52103daee5db53`
MD5	`13fa416ef0fcd1f3384f0690dda16f11`
BLAKE2b-256	`034b49a2c2b0a329737360dc7c7143671d363224508435bf79e79b2c7c4cb041`

Algorithm	Hash digest
SHA256	`a78b0f5e6ebe312c5755df18bf6665a9e3f18c2cb68b780f3a3dc7a050b60b18`
MD5	`2c3fc5b68cae8c5075bee0c916befd2f`
BLAKE2b-256	`78710f26b886bac6af9fbe6a2d430569858e3952b1f9fbea4fe919e0d8a23bc0`

Algorithm	Hash digest
SHA256	`e7d9609d1a7f0bff54502b10c3fdcc01f8dd8e960ef5111ba70171bb2da9231a`
MD5	`99485043fb36974663f828910126a9e8`
BLAKE2b-256	`fe05be358f16a66e8dfd308332ac5b67b68634814adf94f270b30cd213dfd95b`