Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Clair3 - Symphonizing pileup and full-alignment for deep learning-based long-read variant calling
| Date | Stars |
|---|---|
| 2026-07-31 | 384 |
| 2026-08-01 | 385 |
| 2026-08-06 | 385 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
15.0
growth rate 0.00%/day
<div align="center">
<a href="https://en.wiktionary.org/wiki/%E7%9C%BC" target="_blank">
<img src="docs/images/clair3_logo.png" width="110" height="90" alt="Clair3">
</a>
<h1>Clair3</h1>
<p><b>Symphonizing pileup and full-alignment for deep-learning-based long-read variant calling</b></p>
<p>
<a href="https://opensource.org/licenses/BSD-3-Clause"><img src="https://img.shields.io/badge/License-BSD%203--Clause-blue.svg" alt="License"></a>
<a href="http://bioconda.github.io/recipes/clair3/README.html"><img src="https://img.shields.io/badge/install%20with-bioconda-brightgreen.svg?style=flat" alt="install with bioconda"></a>
<a href="https://hub.docker.com/r/hkubal/clair3"><img src="https://img.shields.io/badge/docker-hkubal%2Fclair3-blue.svg" alt="Docker"></a>
</p>
</div>
---
**Contact:** Ruibang Luo, Zhenxian Zheng, Xian Yu
**Email:** [email protected] · [email protected] · [email protected]
---
## Introduction
**Clair3** is a germline small-variant caller for long-read sequencing. It combines two complementary models to balance speed and accuracy:
- **Pileup calling** — fast, handles the majority of variant candidates from summarized alignment statistics.
- **Full-alignment calling** — computationally intensive, resolves uncertain candidates from haplotype-resolved full alignments.
Clair3 is the 3rd generation of [Clair](https://github.com/HKU-BAL/Clair) (2nd) and [Clairvoyante](https://github.com/aquaskyline/Clairvoyante) (1st).
### Looking for a different variant caller?
| Use case | Tool |
| --- | --- |
| Germline on **long-read RNA-seq** | [Clair3-RNA](https://github.com/HKU-BAL/Clair3-RNA) |
| Somatic, **paired tumor/normal** | [ClairS](https://github.com/HKU-BAL/ClairS) |
| Somatic, **tumor-only** | [ClairS-TO](https://github.com/HKU-BAL/ClairS-TO) |
### Agent Skill
[Clair-skills](https://github.com/HKU-BAL/Clair-skills) is a plug-in for agentic AI coding assistants (Claude Code, Cursor, Codex, …) that covers the entire Clair suite. It helps the agent pick the right tool and model, generate ready-to-run commands, and analyze results.
---
## Contents
- [Latest Updates](#latest-updates)
- [Installation](#installation) — [Docker](#option-1-docker) · [Singularity](#option-2-singularity) · [Bioconda](#option-3-bioconda) · [Step-by-step (Conda)](#option-4-step-by-step-conda)
- [Pre-trained Models](#pre-trained-models)
- [Quick Demo](#quick-demo)
- [Usage](#usage)
- [Advanced Topics](#advanced-topics) — [Dwelling time](#dwelling-time-feature) · [Amplicon data](#dealing-with-amplicon-data) · [Postprocessing](#postprocessing-scripts)
- [Reference](#reference) — [Folder structure](#folder-structure-and-submodules) · [Training data](#training-data) · [VCF/GVCF formats](#vcfgvcf-output-formats) · [Model training guides](#model-training-guides)
- [Citation](#citation)
---
## Latest Updates
### v2.0.2 — *Jun 25, 2026*
- Always emit a valid, indexed VCF/gVCF when no variants are found ([Issue#447](https://github.com/HKU-BAL/Clair3/issues/447)).
- Fail early with clear guidance, instead of crashing, when a move-table `*_with_mv` model is run without `--enable_dwell_time` ([Issue#437](https://github.com/HKU-BAL/Clair3/issues/437)).
- Use the longphase conda package on all platforms (longphase now supports linux-aarch64) instead of compiling it at build time ([PR#64876](https://github.com/bioconda/bioconda-recipes/pull/64876)).
- Re-bundle pre-trained models (single pre-packaged archive extracted to \${PREFIX}/bin/models/), restoring bundled models removed since v2.0.0 ([Issue#446](https://github.com/HKU-BAL/Clair3/issues/446)).
### v2.0.1 — *Apr 27, 2026*
- Added the ONT `r1041_e82_400bps_sup_v520_with_mv` signal-aware (move-table) model for Dorado v5.2 SUP basecalled data ([#428](https://github.com/HKU-BAL/Clair3/issues/428)).
- Added a pre-built **GPU Docker image** `hkubal/clair3:v2.0.1_gpu` (CUDA 12.1, PyTorch). See [GPU (NVIDIA CUDA on Linux)](#gpu-nvidia-cuda-on-linux) ([#433](hExcerpt of 40,106 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:13d76dbc050a00aa, topic:deep-learning