Ethical Stack.

HunyuanVideo 1.5

8.3B text-to-video and image-to-video family with 480p/720p checkpoints, step-distilled variants and official inference code.

Score calculation

(3 + 2 + 6 + 1 + 8) ÷ 5 = 4.0

5 of six dimensions rated. Unrated dimensions are excluded from the average.

Full scoring method ↗
Setup details

HunyuanVideo-1.5 official checkpoints and code under Tencent Hunyuan Community License; T2V, I2V and distilled checkpoints have separate runtime settings.

What fits your needs

HunyuanVideo-1.5 official checkpoints and code under Tencent Hunyuan Community License; T2V, I2V and distilled checkpoints have separate runtime settings.

Text and image to video

Official repository lists T2V and I2V checkpoints at 480p and 720p, including selected distilled variants.

See the evidence

Local generation

The repository includes inference code, checkpoint download steps, Diffusers support and a training/finetuning pipeline.

See the evidence

Reduced-step option

A 480p I2V step-distilled release supports 8 or 12 steps; four steps is documented as faster with slightly reduced quality.

See the evidence
Tradeoffs to expect

Territorial license limits

Tencent Community License excludes EU, UK and South Korea and imposes use and redistribution terms.

See the evidence

Prompt rewrite route

The repository supports optional prompt rewriting through a separately configured vLLM endpoint; no endpoint means generation runs without remote rewriting.

See the evidence
How to switch

Download and choose checkpoint

Download the matching T2V or I2V checkpoint and follow the repository command for the target resolution; configure prompt rewriting only if intended.

See the evidence
Practical questions

License territory

The Hunyuan Community License excludes use, reproduction, modification, distribution or display in the EU, UK and South Korea.

See the evidence

Consent and impersonation

The acceptable-use policy prohibits impersonating a person without consent, authorization or legal right and requires clear labeling of generated media in public contexts.

See the evidence

Hosted rewrite

Remote text rewriting is optional; configure a local vLLM endpoint or disable rewrite to avoid sending prompts to an external rewrite service.

See the evidence

Energy practices

3/10↑

This grades disclosed operating practices, not the electricity or emissions of a particular generation.

See the evidence

The release gives limited runtime or control details but no comparable per-task footprint or electricity-source evidence. Tencent publishes an 8.3B model and documented 4090 inference examples for a step-distilled variant, which defines a scoped runtime boundary. The repository offers step-distilled inference, cache options and resolution choices; these are documented runtime controls, not measured physical-energy reductions.

1 + 1 reporting + 0 electricity + 1 controls = 3/10. This grades documented practices and transparency.

Per-task energy, serving region, facility electricity mix, water and independent footprint review are not established in the cited release sources.

Reporting1/3
Tencent publishes an 8.3B model and documented 4090 inference examples for a step-distilled variant, which defines a scoped runtime boundary.
Electricity sourcing0/3
The release materials do not report matched regional electricity sourcing or carbon-free supply for training or inference.
Resource controls1/3
The repository offers step-distilled inference, cache options and resolution choices; these are documented runtime controls, not measured physical-energy reductions.

Water disclosure

Water reporting has not been assessed for this choice.

Training data

2/10↑

The model release does not provide a complete inspectable training-source permission record.

See the evidence

The reviewed release documentation the release describes curation and publishes training code, but does not identify its original training data or source permissions. It provides no source-level permission ledger or creator remedy; a model-weight license governs downstream use, not the material used to train the model.

1 base + 1 source transparency + 0 documented permission + 0 creator controls = 2/10.

Named source inventory, source-level licenses, creator remedies, and independent verification remain unestablished for this release.

Source transparency1/3
The release describes curation and publishes training code, but does not identify its original training data or source permissions.
Documented permission0/4
The release sources do not publish a training-source permission basis or rights ledger; the weight license doesn’t settle training-data permission.
Creator controls0/2
No training-source creator permission-before-use, opt-out, removal or retraining route is documented for this release.
Data-rights rubric ↗

Your data & privacy

—Not enough evidence↑

Privacy has no single score for these downloadable weights because handling depends on the inference host and integrations.

See the evidence

The release can be run locally, but it also appears in hosted provider routes; those operators set their own request retention, training and region terms. A cloud API, where offered, is a separate policy route and is not inherited from the local weights.

A selected host, network path, telemetry, logging, external prompt enhancement, and data retention need route-specific review.

Openness

6/10↑

Weights, inference code and a training pipeline are public under a territorial Community License; training data sources are not published.

See the evidence

Tencent releases model weights, inference and training code and detailed task configuration; the license carries territory and acceptable-use restrictions. The training data inventory is not included. This rating concerns release artifacts and reproducibility, not training-data permission.

6/10 reflects public weights and implementation/training code, with a restrictive regional license and no training data.

Named training sources, source permissions and complete release-specific training data details are not established.

Worker protections

1/10↑

No model-specific worker safeguards are established for Tencent in the reviewed model sources.

See the evidence

The model release pages describe weights and technical properties, not conditions for employees, data annotators, evaluators, moderators, or infrastructure contractors. This is a source-bounded disclosure gap, not evidence of mistreatment. These findings concern the developer or operator. This product’s use in a particular deployment needs separate evidence.

1 base + 0 conditions + 0 worker voice/remedy + 0 evidence/follow-through = 1/10; this is a disclosure finding, not proof of abuse.

Worker locations, pay and conditions, supplier coverage, voice mechanisms, labor audits, and remediation outcomes are not established by these model-release sources.

Pay & conditions0/3
No worker-level conditions or supply-chain safeguards are established in this model-family source set. This is a disclosure gap, not proof of mistreatment.
Worker voice & remedy0/3
The model sources do not document a worker voice or remedy mechanism for data, evaluation, or infrastructure labor.
Evidence & follow-through0/3
No independent worker-level audit or remediation outcome is established in the reviewed model sources.
Workers rubric ↗

Military ties

8/10Not established↑

This finding concerns binding use restrictions on the named release, not proof of how any user deploys it.

See the evidence

The binding Hunyuan license prohibits military-purpose use, but doesn’t spell out a comprehensive ban covering weapons, harmful surveillance and military intelligence; enforcement outcomes are not documented. The reviewed release sources do not document product-specific enforcement outcomes.

7 base + 1 safeguard adjustment = 8/10 (3 of 9 safeguard points; every 3 adds 1). No military relationship was established in the cited sources. This ranks above known defense ties without claiming a verified clean record.

Actual deployments, end users and independent enforcement outcomes are not established by the cited model-release sources.

Use restrictions3/4
The binding Hunyuan license prohibits military-purpose use, but doesn’t spell out a comprehensive ban covering weapons, harmful surveillance and military intelligence; enforcement outcomes are not documented.
Enforcement & oversight0/3
No military-specific review, enforcement or remedy results are documented in the model release materials.
Relationship disclosure0/2
The release documents license terms but no product-specific government or defense deployment.

Release-specific source review found no qualifying model deployment evidence.

Military rubric ↗

BDS targets

Not listed↑

Tencent is not named in the two official BNC campaign lists checked. This is a limited list check, not BDS approval.

See the evidence

The official BNC consumer boycott/pressure guide and No Tech campaign page do not name Tencent among their current targets. BDS uses selective campaigns, so absence from these lists is neutral and should not be read as a clean record or approval. These findings concern the developer or operator. This product’s use in a particular deployment needs separate evidence.

This check covers only named targets in the two cited BNC campaigns; it does not cover all BDS campaigns, local groups, investments, suppliers, or downstream hosts. Lists and ownership may change.

More from Tencent

All sources

Side by side