Skip to main content

haru

The PyTorch implementation of Haru, a full-duplex spoken dialogue model, with a web demo server and an offline inference command.

pip install -e .                  # from this directory
pip install -e ".[webrtc]"        # + WebRTC transport for the web demo
pip install -e ".[triton]"        # + faster audio codec on NVIDIA GPUs
pip install -e ".[share]"         # + public link for the web demo (--share)
haru-server --hf-repo higurashi-ai/Haru-4B       # web demo on http://localhost:8998
haru-inference user.wav reply.wav --reference voice.wav --system-prompt "..."

Requirements: Python 3.10 or later and a CUDA GPU with about 12 GB of memory. ffmpeg on the PATH is needed to use voice clips recorded in the browser (WebM / MP4); WAV, FLAC and Ogg files work without it.

Faster audio codec

With the triton extra, the codec can run as Triton kernels replayed from CUDA graphs: about 0.7 ms instead of 6 ms per encoded or decoded frame on a recent GPU, with the same output up to bfloat16 rounding. It is opt-in:

session = DuplexSession(model, system_prompt=card, reference=voice, config=DuplexConfig(codec_backend="triton"))

The first use in a new environment compiles the kernels (about 15 seconds).

Layout

Module Contents
haru.models.haru HaruModel: language model, audio projections, audio head, codec
haru.models.meanflow_head MeanFlowHead: one audio latent from one hidden state
haru.models.vibevoice_codec VibeVoiceCodec: streaming audio encoder and decoder
haru.codec_triton The same codec as Triton kernels replayed from CUDA graphs
haru.session DuplexSession: a live conversation, one frame at a time
haru.prompt Character cards and the prompt layout
haru.sampling Text sampling, repetition penalty, incremental detokenization
haru.server, haru.webrtc, haru.backends The web demo
haru.run_inference Offline: a recording in, the model's side of the conversation out

See the repository README for the Python API.

Tests

pip install -e ".[dev]"
pytest                                  # CPU-only tests
HARU_MODEL=/path/to/Haru-4B pytest      # + tests that load the checkpoint on a GPU

Metadata

Release files for haru-speech 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for haru-speech 0.1.0
File Size Uploaded Provenance Actions
haru_speech-0.1.0.tar.gz 465.5 kB With provenance Details for haru_speech-0.1.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for haru-speech 0.1.0
File Interpreter ABI Platform Actions
haru_speech-0.1.0-py3-none-any.whl Python 3 none any Details for haru_speech-0.1.0-py3-none-any.whl

Total release size: 926.8 kB

Release files / haru_speech-0.1.0.tar.gz

Download URL haru_speech-0.1.0.tar.gz
Size 465.5 kB
Tags Source
SHA-256 checksum
How to use checksums
4b1fe6713ac102fc08191ed3059fff1177e0a15f4363bc7678299cc672de1656
BLAKE2b-256 checksum
How to use checksums
369c7aafab6abfae50059d04873f20547282fda2d5759efa7355eb96edb1aad7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 13, 2026.

Transparency log

Release files / haru_speech-0.1.0-py3-none-any.whl

Download URL haru_speech-0.1.0-py3-none-any.whl
Size 461.3 kB
Tags Python 3
SHA-256 checksum
How to use checksums
2b8894311ab6921486ecab549f83b933bbf6b36d66b3e3f512190d5f7347e021
BLAKE2b-256 checksum
How to use checksums
ed0120568d0bea88e795d0573480a5b88a16e239812470684d35b8a3429765a5
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 13, 2026.

Transparency log

Security for haru-speech 0.1.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for haru_speech-0.1.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 13, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page