Configuration

scvi_embed · hvg, unscaled

scVI models single-cell RNA counts with a probabilistic latent-variable model that accounts for observed covariates.

Sources (2)

scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations

1 evaluation · 12 results

How it worksscVI workflow

Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.

Sources (2)

scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations

Overview

Model type

Variational autoencoder for count data

Inputs

Cell-by-gene count matrix, optionally with batch, donor or other covariates.

Outputs

Low-dimensional cell representations, normalized expression and probabilistic downstream quantities.

Access

Official project documentation and implementation: https://github.com/scverse/scvi-tools

Sources (2)

scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations

Source reviewed · Automated source review, 2026-09-16. All specifications and missing details

Evaluations and results

1 evaluation · 12 results. Different protocols are not a single leaderboard.

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.949 ARI_cluster/label score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 3 (ARI_cluster/label)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.437 kBET score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 11 (kBET)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	1 cLISI score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 13 (cLISI)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.0986 isolated_label_F1 score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 8 (isolated_label_F1)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.486 cell_cycle_conservation score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 7 (cell_cycle_conservation)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.93 PCR_batch score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 6 (PCR_batch)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.304 iLISI score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 12 (iLISI)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.864 ASW_label/batch score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 5 (ASW_label/batch)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.914 NMI_cluster/label score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 2 (NMI_cluster/label)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.991 graph_conn score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 10 (graph_conn)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.584 isolated_label_silhouette score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 9 (isolated_label_silhouette)
Configuration: scvi_embed · hvg, unscaled	Protocol: scIB official RNA metrics export · pancreas Dataset subset: pancreas	0.596 ASW_label score · higher Uncertainty: status: unreported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source scvi_embed · scIB official RNA metrics export · pancreas scIB official RNA metrics export Aggregation: unreported scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · data/metrics.csv row 230, column 4 (ASW_label)

Source checking is not independent reproduction. Release 2026-09-29-06401fd5b220.

Use this model

How it works, versions and access

Related profile: scVI. This page retains the exact record and its evaluation context.

This configuration

Exact source-reported configuration or submission label. Family membership, checkpoint identity and aliases have not been inferred.

record: scvi_embed · hvg, unscaled
configuration: Not reported
entity type: Configuration

How it works

scVI models single-cell RNA counts with a probabilistic latent-variable model that accounts for observed covariates. Variational autoencoder with a count likelihood and neural encoder/decoder; likelihood and batch/dispersion settings are configurable. The documented inputs are cell-by-gene count matrix, optionally with batch, donor or other covariates. The output consists of low-dimensional cell representations, normalized expression and probabilistic downstream quantities.

Sources (2)

scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations

Versions and reproducibility

scVI model within scvi-tools; package version, likelihood, covariates and checkpoint are evaluation-specific. Gene-feature matrix rather than a fixed sequence-token window.

Sources (2)

scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths supported by sources

Models count observations directly and supports batch-conditioned expression estimates and reference-to-query transfer.
Sources (2)
scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations

Limitations and conditions

The documentation notes that the latent space is less interpretable than a linear method and efficient inference generally benefits from a GPU. Covariates and likelihood must be reported.
Sources (2)
scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations

Profile review details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Stable record: catalog-model-scvi

Specifications

Inputs, training, access and other details

Explanatory profile: source reviewed · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.

Inputs, outputs and configuration
Property	Description and evidence
Model type	Variational autoencoder for count data Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Architecture	Variational autoencoder with a count likelihood and neural encoder/decoder; likelihood and batch/dispersion settings are configurable. Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Inputs	Cell-by-gene count matrix, optionally with batch, donor or other covariates. Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Outputs	Low-dimensional cell representations, normalized expression and probabilistic downstream quantities. Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Parameters	Configuration-dependent, including gene count and encoder/decoder dimensions. Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Known versions	scVI model within scvi-tools; package version, likelihood, covariates and checkpoint are evaluation-specific. Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Training data	Fitted to the user-selected count matrix or a specified pretrained reference; scVI is not one universal checkpoint. Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Training cutoff	Inapplicable as one universal pretraining date: scVI is fitted to the supplied dataset, whose collection date and train/test split belong to the evaluation. · Not applicable Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Context limits	Gene-feature matrix rather than a fixed sequence-token window. Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Weights licence	No universal weights release applies to a model fitted on each dataset; any reused checkpoint requires its own licence. Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Access	Official project documentation and implementation: https://github.com/scverse/scvi-tools Sources (2) scverse/scvi-tools: README.md; scverse/scvi-tools: docs/user_guide/models/scvi.md · docs/user_guide/models/scvi.md: Preliminaries, Generative process, Inference and Limitations
Code licence	BSD-3-Clause Sources scverse/scvi-tools: LICENSE · LICENSE: licence text

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

4 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-29-06401fd5b220
Property and statement	Original source and location	Review and provenance
Relationship: family catalog-model-scvi Individual claims	scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f Original source ↗ scIB pinned metrics.csv method scvi_embed; visualization/exampleKeys.R scVI_embed and scVI_embed_HVG_unscaled; website/R/load.R mapping scVI to scvi; website/R/load.R lines 91-93, 106-114 and 175-180 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 3afbffd3674726e5146797be21cf6bd7470a2c5f Retrieved: 2026-09-19	source checked automated source review · 2026-09-23 Audit details Source review establishes this relationship only. Exact evaluated configurations and original numerical review status remain unchanged. The official benchmark maps the scVI method identity and embedding/full-feature/HVG variants. Keep full_feature versus hvg, unscaled inputs and source version distinct; this establishes method family, not fitted weights. Field: `links:family:catalog-model-scvi` Claim: model-evaluation-identity-0bac8efc1d3ee0b8285e Source artifact SHA-256: `d130cca3688cfe192e2a6df8fa52a0b345d099e259f79014ccf6b90442e626b6` Hash scope: Exact downloaded bytes, before optional gzip storage Inspected artifact
Relationship: family catalog-model-scvi Individual claims	scIB official method labels at 3afbffd Original source ↗ scIB pinned metrics.csv method scvi_embed; visualization/exampleKeys.R scVI_embed and scVI_embed_HVG_unscaled; website/R/load.R mapping scVI to scvi; website/R/load.R lines 91-93, 106-114 and 175-180 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 3afbffd3674726e5146797be21cf6bd7470a2c5f Retrieved: 2026-09-23T11:15:04.764559+00:00	source checked automated source review · 2026-09-23 Audit details Source review establishes this relationship only. Exact evaluated configurations and original numerical review status remain unchanged. The official benchmark maps the scVI method identity and embedding/full-feature/HVG variants. Keep full_feature versus hvg, unscaled inputs and source version distinct; this establishes method family, not fitted weights. Field: `links:family:catalog-model-scvi` Claim: model-evaluation-identity-0bac8efc1d3ee0b8285e Source artifact SHA-256: `f518ee8243adf961b2b2cd99032874c9c304c981c6b36b85d44dac8c2ce2e4ab` Hash scope: Hash scope not separately documented; inspect source record
Relationship: family catalog-model-scvi Individual claims	scIB official method/output decoding at 3afbffd Original source ↗ scIB pinned metrics.csv method scvi_embed; visualization/exampleKeys.R scVI_embed and scVI_embed_HVG_unscaled; website/R/load.R mapping scVI to scvi; website/R/load.R lines 91-93, 106-114 and 175-180 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 3afbffd3674726e5146797be21cf6bd7470a2c5f Retrieved: 2026-09-23T11:15:04.253617+00:00	source checked automated source review · 2026-09-23 Audit details Source review establishes this relationship only. Exact evaluated configurations and original numerical review status remain unchanged. The official benchmark maps the scVI method identity and embedding/full-feature/HVG variants. Keep full_feature versus hvg, unscaled inputs and source version distinct; this establishes method family, not fitted weights. Field: `links:family:catalog-model-scvi` Claim: model-evaluation-identity-0bac8efc1d3ee0b8285e Source artifact SHA-256: `87a0b609ab85f4d637d326a30d1fb5f55ba4dba4173e6c1c870cbf0651d7bcd6` Hash scope: Hash scope not separately documented; inspect source record
Relationship: family catalog-model-scvi Individual claims	scverse/scvi-tools: docs/user_guide/models/scvi.md Original source ↗ scIB pinned metrics.csv method scvi_embed; visualization/exampleKeys.R scVI_embed and scVI_embed_HVG_unscaled; website/R/load.R mapping scVI to scvi; website/R/load.R lines 91-93, 106-114 and 175-180 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 73b28e44223621470e582a81a102c107bb22678b Retrieved: 2026-09-16T19:46:20.137176+00:00	source checked automated source review · 2026-09-23 Audit details Source review establishes this relationship only. Exact evaluated configurations and original numerical review status remain unchanged. The official benchmark maps the scVI method identity and embedding/full-feature/HVG variants. Keep full_feature versus hvg, unscaled inputs and source version distinct; this establishes method family, not fitted weights. Field: `links:family:catalog-model-scvi` Claim: model-evaluation-identity-0bac8efc1d3ee0b8285e Source artifact SHA-256: `c81a5cfb30db1eca8c03af3d0db1a36a447a1bb5a27b270de5633ba73ebeeeef` Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact Inspected artifact

All evidence (CSV)All evidence (JSONL)How to read the evidence

Sources and history

View linked audit checks and correction history

Release 2026-09-29-06401fd5b220 · Record review: discovered

1 source records and release history

scIB: 3afbffd3674726e5146797be21cf6bd7470a2c5f · Original source · 3afbffd3674726e5146797be21cf6bd7470a2c5f

Download this release

Technical metadata and extraction receipts

Stable ID: acquired-configuration-ad6e8cb9486fa77b3173

source locator: data/metrics.csv row 230, column 2 (NMI_cluster/label)
reported configuration: scvi_embed
reference kind: not_a_reference
missing metadata: checkpoint revision: unreported; model family: unextracted

Related records

family: scVI
model: scvi_embed · scIB official RNA metrics export · pancreas
subject: scvi_embed · hvg, unscaled: family scVI

Suggest a correction