whatmodel.sh · find the best video models you can run

Best text-to-video models

A leaderboard of the best text-to-video models. Gemini Omni 1.1 Flash is #1 right now, followed by Gemini Omni Flash and FLUX 3 Video. Models are ranked on human votes and video generation tests. Filter by your computer or budget to find the best one you can use.

97 models · scores combine 25 benchmarks · updated 2026-10-11 02:20 UTChow it works · find a leaderboard · benchmark your machine

task[all tasks][text to video][image to video][video editing]

Scores use reviewed text to video evaluations only. Compare scores within this task. Models without task measurements are excluded.

Filter by your hardware and budget · reset

+ more

set[all sets][LMArena video · 94][VABench T2AV micro Visual QA · 3]

Filter the model list while keeping task scores. Models without a shared score scale are not yet calibrated; select a set to see its own scores.

Detect your exact chip or graphics card:

curl -G whatmodel.sh/video -d max=0 --data-urlencode "mac=$(sysctl -n hw.memsize machdep.cpu.brand_string)"Mac

curl -G whatmodel.sh/video -d max=0 --data-urlencode "ram=$(awk '/MemT/{print $2}' /proc/meminfo)" --data-urlencode "nv=$(nvidia-smi --query-gpu=name,memory.total --format=csv,noheader 2>/dev/null)"Linux

curl.exe -G whatmodel.sh/video -d max=0 --data-urlencode "ram=$((gcim Win32_ComputerSystem).TotalPhysicalMemory)" --data-urlencode "nv=$(nvidia-smi --query-gpu=name,memory.total --format=csv,noheader 2>$null)"Windows

$ terminal

ram[8][16][24][32][48][64][96][128][192][256]GB

gpu[apple silicon][graphics card][none]

price[local][$0.02][$0.05][$0.1][$0.25][any]max per sec

detecting your OS…

Score vs API price

  80 ┤ · · · · · · · · · · · · · · · · ·Gemini Omni 1.1 Flash·◆· · · · · · · ·◇· · · · · · · · · · · · · · · ·
     │   ·           Grok Imagine Video·             Wan 3.0 ◆┘       ∘ ◇           ·            ◇        ◇
  60 ┤ · ◆─────────────────────────◆◇─◆◆◇─────────◇──────────◇ · ◇ ◇ · · · · · · · · · · · · · · · · · · · · ·
     │   ∘                     ∘    ◇  ·◇    ◇      ∘ ∘◇     ·                      ·     ∘    ◇
  40 ┤ · · · · · · · · · · · · ◇ · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ·
     └───┬─────────────────────────────┬─────────────────────┬──────────────────────┬─────────────────────────
       $0.02                         $0.05                 $0.1                   $0.2

#	model	task score↓	where↓	speed↓	RAM↓	get it	sources

#	model	task score↓	where↓	RAM↓	get it
1	Gemini Omni 1.1 Flash Google	82±2*	api $0.10/sec		ai.google.dev ↗
2	Gemini Omni Flash Google	81±2*	api $0.10/sec		ai.google.dev ↗
3	FLUX 3 Video Black Forest Labs	80±2*	api $0.17/sec		docs.bfl.ai ↗
4	Dreamina Seedance 2.0 ByteDance Seed	78±2*	api $0.30/sec		fal.ai ↗
5	Wan 3.0 Alibaba	78±2*	api $0.10/sec		www.alibabacloud.com ↗
6	Dreamina Seedance 2.5 ByteDance	78±2*	api $0.30/sec		docs.dev.runwayml.com ↗
7	Muse Video Meta	77±2*	comparison only
8	MiniMax H3 MiniMax	76±2*	api $0.13/sec +local	~40GB	fal.ai ↗
9	HappyHorse 1.1 Alibaba	75±7*	api $0.14/sec		www.alibabacloud.com ↗
10	HappyHorse 1.0 Alibaba-ATH	74±2*	api $0.14/sec		replicate.com ↗
11	Sora 2 Pro OpenAI	70±2*	comparison only
12	Veo 3.1 Google	69±2	api $0.40/sec		ai.google.dev ↗
13	Veo 3.1 Fast Google	69±2	api $0.10/sec		ai.google.dev ↗
14	Veo 3 Fast Google	68±2	comparison only
15	Grok Imagine Video xAI	68±2*	api $0.050/sec		docs.x.ai ↗
16	Veo 3 Google	67±2*	comparison only
17	Wan 2.7 Alibaba	67±2*	api $0.10/sec		www.alibabacloud.com ↗
18	Sora 2 OpenAI	67±2	comparison only
19	Wan 2.6 Alibaba	66±2	api $0.10/sec		www.alibabacloud.com ↗
20	Seedance 1.0 Pro Fast ByteDance	63±10*	api $0.049/sec		fal.ai ↗
21	Seedance 1.5 Pro ByteDance Seed	61±2*	api $0.052/sec		fal.ai ↗
22	Wan 2.5 Preview Alibaba	60±2*	api $0.10/sec		www.alibabacloud.com ↗
23	PixVerse V5.6 PixVerse	59±2*	api $0.045/sec		docs.platform.pixverse.ai ↗
24	Runway Gen-4.5 Runway	58±2*	api $0.12/sec		docs.dev.runwayml.com ↗
25	Kling 2.5 Turbo Pro Kuaishou	58±2*	api $0.070/sec		fal.ai ↗
26	Kling 2.6 Pro Kuaishou (Kling AI)	58±2*	api $0.070/sec		fal.ai ↗
27	Cosmos3-Super NVIDIA	58±5*†	local	~160GB	how to run ›
28	Ray 3 Luma AI	57±2*	comparison only
29	P-Video Pruna	57±2*	api $0.020/sec		replicate.com ↗
30	Hailuo 2.3 MiniMax	57±2*	api $0.047/sec		platform.minimax.io ↗
31	Kling O1 Pro Kuaishou (Kling AI)	57±2*	api $0.11/sec		replicate.com ↗
32	Seedance 1.0 Pro ByteDance Seed	56±2*	api $0.052/sec		docs.byteplus.com ↗
33	Hailuo 02 MiniMax	55±2*	api $0.047/sec		platform.minimax.io ↗
34	Kandinsky 5.0 Pro Sber	54±2*	api $0.040/sec +local	~76GB	fal.ai ↗
35	Cosmos3-Nano NVIDIA	54±4*	local	~42GB	how to run ›
36	HunyuanVideo 1.5 Tencent	54±2	api $0.075/sec +local	~20GB	fal.ai ↗
37	Veo 2 Google	53±2*	comparison only
38	PixVerse 5.5 PixVerse	53±9*	api $0.060/sec		docs.platform.pixverse.ai ↗
39	Kling 2.1 Master Kuaishou (Kling AI)	53±2*	api $0.28/sec		replicate.com ↗
40	LongCat-Video Meituan	53±7*	api $0.040/sec +local	~100GB	fal.ai ↗
41	LTX-2 Lightricks	53±2*	local	~120GB	how to run ›
42	Wan 2.2 A14B Alibaba	52±2	api $0.080/sec +local	~31GB	fal.ai ↗
43	Cosmos-1.0-Diffusion-7B-Video2World NVIDIA	51±5*†	comparison only
44	Wan 2.1 14B Alibaba	50±2	api $0.24/sec +local	~20GB	replicate.com ↗
45	Seedance 1.0 Lite ByteDance Seed	50±2*	comparison only
46	Kandinsky 5.0 T2V Lite Sber	49±2*	local	~29GB	how to run ›
47	LTX-2 19B Distilled Lightricks	49±7*	api $0.020/sec +local	~41GB	replicate.com ↗
48	Cosmos-1.0-Diffusion-7B-Text2World NVIDIA	48±3	local	~30GB	how to run ›
49	Cosmos-Predict2.5-14B NVIDIA	48±5*†	comparison only
50	Hailuo 01 (MiniMax video-01 / I2V-01) MiniMax	47±5*	api $0.083/sec		fal.ai ↗
51	Cosmos-Predict2.5-2B NVIDIA	47±5*†	comparison only
52	Wan 2.2 TI2V 5B Alibaba	47±4	api price unknown +local	~6.5GB	fal.ai ↗
53	CogVideoX-5B Zhipu AI (THUDM)	47±2	api price unknown +local	~13GB	fal.ai ↗
54	Kling 1.6 (Standard mode) Kuaishou (Kling AI)	47±5*	comparison only
55	Ray 2 Luma AI	45±2*	comparison only
56	Sora (original, 2024) OpenAI	45±2	comparison only
57	Pyramid Flow (SD3) Peking University / Kuaishou	45±3*	local	~8GB	how to run ›
58	T2V-Turbo-v2 UC Santa Barbara	45±7*†	api price unknown +local	~9.4GB	replicate.com ↗
59	Open-Sora 2.0 HPC-AI Tech	45±5*	local	~55GB	how to run ›
60	Luma Dream Machine 1.0 Luma AI	43±6*	comparison only
61	Wan 2.1 T2V 1.3B Alibaba	43±4	api $0.040/sec +local	~8.6GB	replicate.com ↗
62	LTX-2.3 Lightricks	43±7*	api $0.040/sec +local	~34GB	docs.ltx.io ↗
63	Mochi 1 Genmo	42±2	api price unknown +local	~6.5GB	replicate.com ↗
64	T2V-Turbo (MS) UC Santa Barbara	42±6*†	comparison only
65	HunyuanVideo Tencent	41±3	api price unknown +local	~28GB	fal.ai ↗
66	Pika 2.2 Pika	41±2	api $0.040/sec		fal.ai ↗
67	Runway Gen-3 Alpha Runway	41±3	comparison only
68	T2V-Turbo (VC2) UC Santa Barbara	41±3†	local	~9.4GB	how to run ›
69	Vidu 1.5 ShengShu (Vidu)	40±3*	comparison only
70	CogVideoX-2B Zhipu AI (THUDM)	39±2	local	~5.3GB	how to run ›
71	CogVideoX1.5-5B Zhipu AI (THUDM)	38±3	local	~38GB	how to run ›
72	Allegro Rhymes AI	38±2	local	~9.3GB	how to run ›
73	VideoCrafter2 Tencent AI Lab	37±2	local	~9.4GB	how to run ›
74	Kling 1.0 Kuaishou (Kling AI)	37±6*	comparison only
75	Cosmos-1.0-Autoregressive-4B NVIDIA	36±5*†	comparison only
76	LaVie (base T2V) Shanghai AI Lab (Vchitect)	35±2	local	~4.9GB	how to run ›
77	LTX-Video 0.9 (2B) Lightricks	35±3*	local	~5.3GB	how to run ›
78	Pika 1.0 Pika	35±3	comparison only
79	Open-Sora 1.1 HPC-AI Tech	34±6*	local	~14GB	how to run ›
80	Cosmos-1.0-Autoregressive-5B-Video2World NVIDIA	33±5*†	comparison only
81	NOVA-0.6B T2V BAAI	32±3*	local	~9.3GB	how to run ›
82	Hotshot-XL Hotshot	31±2	local	~10GB	how to run ›
83	PikaLab (September 2023) Pika	30±2	comparison only
84	Runway Gen-2 Runway	30±2	comparison only
85	ZeroScope v2 576w cerspense	30±3	local	~7.9GB	how to run ›
86	Lumiere (T2V) Google	30±4*	comparison only
87	VideoCrafter1 T2V Tencent AI Lab / HKUST	29±2	local	~8.1GB	how to run ›
88	Show-1 Show Lab / NUS	29±2	local	~83GB	how to run ›
89	Open-Sora 1.2 HPC-AI Tech	28±3*	local	~39GB	how to run ›
90	ModelScope T2V 1.7B Alibaba DAMO Academy	28±2	local	~4.6GB	how to run ›
91	LVDM (text-to-video) HKUST / Tencent AI Lab	28±3*	local	~6.1GB	how to run ›
92	Floor33 Pictures (August 2023) Floor33 Pictures	27±2	comparison only
93	ZeroScope v2 XL (576w → XL pipeline) cerspense	27±2	local	~15GB	how to run ›
94	CogVideo 9B (original) Tsinghua / Zhipu	26±7*	local	~52GB	how to run ›
—	Kandinsky 6.0 Video Lite Sber	not yet calibrated	local	~34GB	how to run ›
—	Kandinsky 6.0 Video Pro Sber	not yet calibrated	local	~109GB	how to run ›
—	LTX-2.5 Lightricks	not yet calibrated	api $0.12/sec +local	~14GB	docs.ltx.io ↗

Includes 26 comparison-only models with no verified API or download · hide them

97 of 97 catalogue models listed here · sources and licences

* new or thinly tested: ranked on too few benchmark results · † only the maker’s own numbers so far · ~RAM is estimated; full context may need more memory