Benchmark

Liquids leaderboard

Board: Overall · Hands · Faces · Liquids · On-screen text · Multi-shot · Lip-sync · Camera · Animals · Style · Image to video · Animation

Look: Realism · Animation and 3D

Round: all · v1-vivix-2026-10 · v1-2026-10 · v1-2026-09 · anim1-2026-10

Clip shown: every model here on the same prompt, “liquid-coffee-pour”, so the preview is a fair comparison, not a cherry-picked best clip.

#	Model	Clip	Overall	95% CI	Fidelity	Visual	Physics	Consistency	Audio	Attempts	Gen (s)	$/run	Runs
1	Grok Video xAI		4.83	±0.53	4.88	4.88	4.75	—	—	1.00	85	—	4
2	Kling 3.0 Turbo Kuaishou		4.81	±0.26	4.75	4.78	4.73	4.78	5.00	—	—	—	4
3	Google Gemini Omni Google		4.79	±0.36	4.75	4.78	4.78	4.83	4.80	—	—	—	4
4	Seedance 2.0 ByteDance		4.63	±1.19	4.88	4.75	4.25	—	—	—	237	—	4
5	Seedance 2.5 ByteDance		4.63	±1.19	4.88	4.75	4.25	—	—	—	233	—	4
6	Kling 3.0 Kuaishou		4.46	±0.76	4.50	4.50	4.38	—	—	1.00	190	—	4
7	Runway Gen-4.5 Runway		4.33	±0.75	4.45	4.40	3.95	4.53	—	—	—	—	4
8	MiniMax H3 MiniMax		4.25	±2.39	4.25	4.50	4.00	—	—	—	448	—	4
9	OpenAI Sora 2 discontinued OpenAI		4.21	±1.46	4.63	4.38	3.63	—	—	1.00	217	—	4
10	Google Veo 3.1 Google		4.21	±2.52	4.25	4.38	4.00	—	—	—	131	—	4
11	MiniMax Hailuo 02 MiniMax		4.17	±2.65	4.13	4.38	4.00	—	—	—	300	—	4
12	PixVerse v5.5 PixVerse		4.13	±0.31	4.33	4.15	3.70	4.35	—	—	—	—	4

Attempts is the average number of generations needed before one was usable, the number that actually decides what a model costs you. 95% CI is computed from that model's own runs on this board, not assumed. On a category board with only 4 runs per model, a wide CI is the honest reflection of a small sample, not a bug. Two models whose CIs overlap are not meaningfully different at this sample size.

Frequently asked

Which AI video model is best at Liquids right now?

As of 23 October 2026, Grok Video by xAI ranks #1 on Liquids on SlopTV's benchmark, scoring 4.83/5 across 4 scored runs (±0.53 at 95% confidence).