Are there independent benchmarks for avatar generation models, or only vendor-/author-reported numbers?

The question

Most performance claims in this space come from the model authors or the vendor selling the model:

• wan-animate reports beating stableanimator, AnimateAnyone, and Champ on automated metrics, and beating runway-act-two and DreamActor-M1 on human eval.
• heygen-avatar-v reports 68.9–85.7% pairwise preference over competitors.
• Closed commercial roundups (Kling 3.0 motion fidelity score 8.4, Aurora "best in class") are vendor-marketing.

Is there an independent benchmarking effort — VBench-class — that compares avatar models on a level playing field?

Why it matters

Without independent benchmarks, picking a model means trusting the vendor or paper authors. For a thread specifically about evaluating model capabilities, this is the load-bearing question.

What we currently believe

VBench has been mentioned as a relevant benchmark for video generation broadly, and DisPose's results report VBench scores (dispose-pose-conditioning). Whether VBench has avatar-specific tasks, or whether there's a dedicated avatar leaderboard, is unknown.

Evidence we have

• 2026-05-07-ai-avatar-motion-mimicking-models-survey flags this gap in its "Contradictions and open questions" section.
• DisPose reports VBench improvements over MimicMotion, suggesting VBench is at least applicable.

Evidence we need

• Whether VBench has dedicated avatar-relevant categories.
• Whether there's a standalone avatar leaderboard (e.g., on HuggingFace or paperswithcode).
• Independent third-party comparisons (academic surveys, benchmarking papers).

How to resolve

• Search VBench documentation for human-animation task definitions.
• Look for 2025 survey papers on "human image animation" or "talking-head generation" benchmarking.
• A targeted /academic-research pass on the topic.

Related

• wan-animate
• heygen-avatar-v
• stableanimator
• hand-fidelity-comparison-across-avatar-models