Benchmark Almanac

EvalVerse

Not in model cards

Video models generate cinematic clips from prompts and reference material derived from professional videos. Film experts compare outputs across visual design, acting, cinematography, aesthetics, emotion, shot sequences, and sound. A calibrated vision-language evaluator also estimates these judgments.

The paper reports both expert comparisons and automated scores. These are different measurements. The opened paper establishes the protocol, but a complete public release of the test assets was not verified.

Released
May 2026
Kind
Video generation
Level
Expert
Scored
Graded by people
Answers
Open-ended
Measures
Preference
Subject
Arts
Language
English
In and out
Text, Video → Video, Audio
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
A host
Scores refreshed
November 20th, 2026

In the authors’ words

The rapid evolution of generative video foundation models has propelled the field toward professional-grade cinematic synthesis. To achieve such demanding quality, the community transitions towards Reinforcement Learning (RL) and agentic workflows. However, reliable evaluation has emerged as a critical bottleneck. Existing benchmarks predominantly evaluate ”whether it is right” (basic prompt-following) while fundamentally neglecting ”whether it is good” (cinematic quality, acting, and aesthetics). Furthermore, current automated metrics lack the domain-specific rigor required to provide trustworthy signals, creating a severe credibility gap between human aesthetic perception and machine scoring. To bridge this gap, we introduce EvalVerse, a comprehensive, pipeline-aware, and expert-calibrated evaluation framework. We treat video generation assessment not merely as an engineering task, but as a core scientific problem: the systematic digitization of subjective cinematic expertise. First, we organize domain knowledge into an evaluation taxonomy aligned with the professional filmmaking workflow (pre-production, production, and post-production). Second, we distill human expert judgments into a curated dataset with large-scale human annotations. Third, we inject this knowledge into Vision-Language Models (VLMs) through an expert-calibrated fine-tuning strategy, enabling the VLM to perform explicit Chain-of-Thought reasoning. Compared to previous works, EvalVerse not only retains compatibility with foundational ”rightness” metrics, but also significantly expands the criteria to ”goodness” and broaden the task coverage to complex multi-shot sequencing and audio-visual integration. Consequently, by providing granular diagnostic signals, EvalVerse transcends a static leaderboard and establishes a fundamental infrastructure for future work, such as reward models and evaluator agent.

Abstract of EvalVerse: Pipeline-Aware and Expert-Calibrated Benchmarking for Professional Cinematic Video Generation (arXiv:2605.23271, CC0)
3
Entries in its paper, 2 models we track

HappyHorse 1.0 leads with 0.73 HappyHorse 1.0 leads with 0.73

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper reports no overall score, so the headline averages the expert pairwise win ratios across all 18 dimensions. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

  1. 1 HappyHorse 1.0 0.73
  2. 2 Kling-v3-Omni 0.68
  3. 3 Dreamina Seedance 2.0 0.68

Source: The EvalVerse paper

Source: The EvalVerse paper

#EntryScoreSettingDateSource
1 HappyHorse 1.0 0.73 Official leaderboard
2 Kling-v3-Omni 0.68 Official leaderboard
3 Dreamina Seedance 2.0 0.68 Official leaderboard
#ModelScoreSettingDateSource
1 HappyHorse 1.0 0.73 Official leaderboard
2 Dreamina Seedance 2.0 0.68 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

No dated values yet.

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Audio and video

All 695 in Audio and video
  • Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy. Qwen 3.8 Max 88.7%
  • LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization. Gemini 4 Argon 91.7%
  • MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy. Qwen 3.8 Max 82.4%
  • MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy. Qwen 3.8 Max 90.8%
  • FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better. Phi-4-multimodal 4.00%
  • VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules. Qwen3.5-Omni-Plus 93.1%

Data audit

6 of 7 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 pass 4 notes
  • Pass Independent runs Scores from the benchmark’s paper (3 entries).
  • Note Models scored Only 2 models we track, plus 1 other entry on its leaderboard.
  • Note Newest score No scores yet.
  • Note Frontier models None of its models are on the capabilities index.
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 2 rows for models we track and 1 other entry at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check