Measured video performance

Video generation performance

Name: TrustedRouter AI video model leaderboard
Creator: TrustedRouter
License: https://www.apache.org/licenses/LICENSE-2.0

Measured completion time, reliability, and cost for asynchronous video models served directly through the attested TrustedRouter gateway.

Generate video Text leaderboard

Last updated 2026-08-01T20:43:12Z

Video jobs are asynchronous, so token throughput and TTFT are the wrong measurements. This leaderboard uses end-to-end completion time, successful completion rate, exact customer cost, and cost per generated second from a rolling video benchmark set of up to 5,000 jobs. Rankings use successful completion first, then p50 completion time and cost. Prompt, reference media, generated bytes, download URLs, API keys, and workspace IDs are never stored in benchmark rows. Video traffic goes directly from the attested gateway to the named provider and never passes through OpenRouter.

Direct providers

Actual provider paths, not model publishers. A model offered through multiple direct providers receives one row per provider.

#	Provider	Models	Completed	p50 total	p95 total	p50 cost / output sec	Jobs
1	grok	1	100.00%	13.4 s	17.5 s	$0.06	2
2	ltx	1	100.00%	21.9 s	21.9 s	$0.072	1
3	kling	1	100.00%	24.9 s	24.9 s	$0.109092	1
4	venice	5	85.71%	27.1 s	180.0 s	$0.04	7
5	runway	1	20.00%	44.1 s	44.1 s	$0.144	5

Video models

Operational measurements are kept separate from visual quality. A fixed public prompt suite and blind quality results will appear only after each model has enough comparable runs.

#	Model	Provider	Inputs	Audio	Completed	p50 total	p95 total	p50 time / output sec	p50 job cost	p50 cost / output sec	Jobs
1	pixverse/c1	venice	text	No	100.00%	11.4 s	38.7 s	3.8 s	$0.12	$0.04	2
2	x-ai/grok-imagine-video	grok	text	No	100.00%	13.4 s	17.5 s	13.4 s	$0.06	$0.06	2
3	lightricks/ltx-2.3-fast	ltx	text	No	100.00%	21.9 s	21.9 s	3.6 s	$0.432	$0.072	1
4	kling/v3-pro	kling	text	No	100.00%	24.9 s	24.9 s	8.3 s	$0.327276	$0.109092	1
5	lightricks/ltx-2.3-fast	venice	—	—	100.00%	27.1 s	27.1 s	—	$0.42	—	1
6	google/gemini-omni-flash	venice	—	—	100.00%	27.1 s	27.1 s	—	$0.5775	—	1
7	bytedance/seedance-2.0-fast	venice	—	—	100.00%	104.2 s	104.2 s	—	$0.798	—	1
8	minimax/hailuo-3	venice	—	—	50.00%	180.0 s	180.0 s	—	$0.8505	—	2
9	runway/gen-4.5	runway	text	No	20.00%	44.1 s	44.1 s	22.1 s	$0.288	$0.144	5

Next measurement layer

Quality without hand waving.

The quality suite will hold prompt, duration, aspect ratio, resolution, and audio settings constant. It will report prompt adherence, motion consistency, visual artifacts, identity consistency, text rendering, and audio synchronization separately.

Blind human preference and machine-judge scores will remain separate from reliability and cost. TrustedRouter will publish the prompts and replay metadata needed to reproduce every score.