Video generation

• Best overallSWan 3.01,174 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)
• Best valueSMiniMax H3Tier S for $0.08/s of video
• Best free startSFLORAFree tier · from $18/mo · via Wan 3.0
• Best local (≤24 GB GPU)CHunyuanVideo 1.5~14 GB VRAM · 1,169 on LMArena (Arena) Text-to-Video leaderboard

• SABCTier: how close to the best result on the task's main public benchmark
• A Outlined: a tool inherits the tier of the best model it offers
• – No public result yet
• Cloud · Local: runs on your own hardware

CloudLocal

Models

The engines themselves. Use them through an API or inside the apps on the right.

Cloud

S Wan 3.0Alibaba · $0.2/s of video1,174 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)$S Utopai XUtopai Studios1,169 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)S Seedance 2.5ByteDance (Seed) · $0.57/s of video1,160 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)S FLUX 3 VideoBlack Forest Labs · $0.29/s of video1,146 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)A BACH 1.0 ProVideo Rebirth · $0.12/s of video1,362 on Artificial Analysis Video Arena: Image to Video, no audio (AA-Video-I2V-Silent v1.0)$A Gemini Omni Flash 1.1Google · $0.1/s of video1,132 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)$B Grok Imagine Video 1.5xAI · $0.08/s of video1,060 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)$B HappyHorse 1.1Alibaba (ATH) · $0.18/s of video1,055 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)C PixVerse V6PixVerse · $0.12/s of video988 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)C Midjourney Video V1Midjourney1,165 on Artificial Analysis Video Arena: Image to Video, no audio (AA-Video-I2V-Silent v1.0)C Veo 3.1Google · $0.4/s of video968 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)C Gen-4.5Runway · $0.12/s of video913 on Artificial Analysis Video Arena: Text to Video, no audio (AA-Video-T2V-Silent v2.1)C Vidu Q4 PreviewShengShu Technology941 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)C Pika 2.5Pika Labs · $0.09/s of video838 on Artificial Analysis Video Arena: Text to Video, no audio (AA-Video-T2V-Silent v2.1)C Ray3.2Luma AI1,206 on LMArena (Arena) Text-to-Video leaderboardC BorealCreatify · $0.01/s of video831 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)$– Kling 4.0Kuaishou (Kling AI)No public result yet– P-Video-2Pruna AI · $0.03/s of videoNo public result yet– PixVerse C1PixVerseNo public result yet

Local

S MiniMax H3MiniMax · Open weights1,159 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)A Cosmos3-Super-Image2VideoNVIDIA · Open weights1,275 on Artificial Analysis Video Arena: Image to Video, no audio (AA-Video-I2V-Silent v1.0)B MAGI-2 PreviewSand.ai · ~640 GB VRAM1,045 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)$C LTX-2.5Lightricks · Open weights950 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)C HunyuanVideo 1.5Tencent · ~14 GB VRAM1,169 on LMArena (Arena) Text-to-Video leaderboard$C Wan 2.2Alibaba (Wan-AI) · ~24 GB VRAM1,132 on LMArena (Arena) Text-to-Video leaderboard– daVinci-MagiHumanSII-GAIR and Sand.ai · Open weightsNo public result yet– HuMo-17BByteDance (Intelligent Creation Lab) · ~24 GB VRAMNo public result yet– Kandinsky 6.0 Video ProKandinsky Lab · Open weightsNo public result yet– LongCat-VideoMeituan (LongCat) · Open weightsNo public result yet– MOVA (360p / 720p)OpenMOSS · ~12 GB VRAMNo public result yet– SkyReels V3Skywork AI (Kunlun Tech) · Open weightsNo public result yet

Apps

Ready-made apps with a UI: sign in or install, then start.

Cloud

S HiggsfieldHiggsfield AIvia Wan 3.0S invideo AIinvideovia Wan 3.0S KreaKrea · Free tier · $5.25/movia Wan 3.0$S Magnific (formerly Freepik)Freepik Company · $20/movia Wan 3.0S PikaPika Labs · Free tier · $10/movia Wan 3.0S RunwayRunway · Free tier · $15/movia Wan 3.0S WanAlibabavia Wan 3.0S PAI (Utopai Studios)Utopai Studios · $15/movia Utopai XS CreatifyCreatify · Free tier · $15/movia Seedance 2.5S DreaminaByteDance (CapCut) · Free tiervia Seedance 2.5S Hailuo AIMiniMaxvia Seedance 2.5S LTX StudioLightricks · Free tier · $15/movia Seedance 2.5S LumaLuma AI · Free tier · $30/movia Seedance 2.5S OpenArtOpenArt · Free tier · $14/movia Seedance 2.5A BACHVideo Rebirth · Free tiervia BACH 1.0 ProA Adobe FireflyAdobe · $9.99/movia Gemini Omni Flash 1.1A FlowGooglevia Gemini Omni Flash 1.1A GeminiGoogle · Free tier · $4.99/movia Gemini Omni Flash 1.1$B GrokSpaceXSI (formerly xAI) · Free tier · $8/movia Grok Imagine Video 1.5C PixVersePixVerse · Free tier · $10/movia PixVerse V6C MidjourneyMidjourney · $10/movia Midjourney Video V1C CanvaCanva · Free tier · $18/movia Veo 3.1C Gemini in Google WorkspaceGooglevia Veo 3.1C Leonardo.AiLeonardo.Ai (Canva) · Free tier · $12/movia Veo 3.1C ViduShengShu Technology · Free tiervia Vidu Q4 Preview– CivitaiCivitai · Free tier · $10/moNo public result yet– DzineDzine · $8.99/moNo public result yet– getimg.aigetimg.ai · $15/moNo public result yet– GlifGlif · Free tierNo public result yet– ImagineArtImagineArt · $5/moNo public result yet– Kling AIKuaishou · Free tierNo public result yet– LovartLovart · Free tierNo public result yet– MageMage · Free tier · $10/moNo public result yet– Meta AIMeta · Free tierNo public result yet– NightCafeNightCafe Studio · Free tierNo public result yet– ScenarioScenario · Free tier · $15/moNo public result yet– SeaArt AISeaArtNo public result yet

Local

S WanGP (Wan2GP)deepbeepmeep (community project) · Freevia MiniMax H3$S SwarmUImcmonkeyprojects · Freevia MiniMax H3$C SD.Nextvladmandic (community project) · Freevia LTX-2.5C LTX DesktopLightricks · Freevia LTX-2.5C Draw ThingsDraw Things · Freevia LTX-2.5C AmuseTensorStack · Freevia Wan 2.2– SD.Nextvladmandic (community) · FreeNo public result yet

Workflow platforms

Node and canvas tools that chain several models into one pipeline.

Cloud

S FLORAFLORA · Free tier · $18/movia Wan 3.0S KreaKrea · Free tier · $5.25/movia Wan 3.0$S Magnific (formerly Freepik)Freepik Company · $20/movia Wan 3.0S Figma WeaveFigma · Free tier · $24/movia Seedance 2.5– Comfy CloudComfy Org · Free tier · $20/moNo public result yet– FuserFuserNo public result yet– RunComfyRunComfyNo public result yet– ScenarioScenario · Free tier · $15/moNo public result yet– ThinkDiffusionThinkDiffusionNo public result yet

Local

S ComfyUIComfy Org · Freevia MiniMax H3$S SwarmUImcmonkeyprojects · Freevia MiniMax H3$

Runtimes & APIs

Run open models on your own machine, or reach models and services through an API.

Cloud

S falfalvia Wan 3.0S ReplicateReplicatevia Wan 3.0– RunwareRunwareNo public result yet– WaveSpeedAIWaveSpeedAINo public result yet

Local

Nothing tracked yet

Everything, in order

All video generation options in quality order

Tier	Name	Approach	Runs	Rated by	Price	Skill
S	Wan 3.0 Alibaba	Models	Cloud	1,174 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.2/s of video	—
S	Utopai X Utopai Studios	Models	Cloud	1,169 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	—	—
S	Seedance 2.5 ByteDance (Seed)	Models	Cloud	1,160 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.57/s of video	—
S	MiniMax H3 MiniMax	Models	Local	1,159 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	Open weights	—
S	FLUX 3 Video Black Forest Labs	Models	Cloud	1,146 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.29/s of video	—
S	fal fal	Runtimes & APIs	Cloud	via Wan 3.0	—	intermediate
S	FLORA FLORA	Workflow platforms	Cloud	via Wan 3.0	Free tier · $18/mo	intermediate
S	Higgsfield Higgsfield AI	Apps	Cloud	via Wan 3.0	—	beginner
S	invideo AI invideo	Apps	Cloud	via Wan 3.0	—	beginner
S	Krea Krea	Apps + Workflow platforms	Cloud	via Wan 3.0	Free tier · $5.25/mo	beginner
S	Magnific (formerly Freepik) Freepik Company	Apps + Workflow platforms	Cloud	via Wan 3.0	$20/mo	beginner
S	Pika Pika Labs	Apps	Cloud	via Wan 3.0	Free tier · $10/mo	beginner
S	Replicate Replicate	Runtimes & APIs	Cloud	via Wan 3.0	—	intermediate
S	Runway Runway	Apps	Cloud	via Wan 3.0	Free tier · $15/mo	beginner
S	Wan Alibaba	Apps	Cloud	via Wan 3.0	—	beginner
S	PAI (Utopai Studios) Utopai Studios	Apps	Cloud	via Utopai X	$15/mo	intermediate
S	Creatify Creatify	Apps	Cloud	via Seedance 2.5	Free tier · $15/mo	beginner
S	Dreamina ByteDance (CapCut)	Apps	Cloud	via Seedance 2.5	Free tier	beginner
S	Figma Weave Figma	Workflow platforms	Cloud	via Seedance 2.5	Free tier · $24/mo	intermediate
S	Hailuo AI MiniMax	Apps	Cloud	via Seedance 2.5	—	beginner
S	LTX Studio Lightricks	Apps	Cloud	via Seedance 2.5	Free tier · $15/mo	beginner
S	Luma Luma AI	Apps	Cloud	via Seedance 2.5	Free tier · $30/mo	beginner
S	OpenArt OpenArt	Apps	Cloud	via Seedance 2.5	Free tier · $14/mo	beginner
S	ComfyUI Comfy Org	Workflow platforms	Local	via MiniMax H3 Cloud models: tier S	Free	advanced
S	WanGP (Wan2GP) deepbeepmeep (community project)	Apps	Local	via MiniMax H3	Free	intermediate
S	SwarmUI mcmonkeyprojects	Apps + Workflow platforms	Local	via MiniMax H3	Free	intermediate
A	BACH 1.0 Pro Video Rebirth	Models	Cloud	1,362 on Artificial Analysis Video Arena: Image to Video, no audio (AA-Video-I2V-Silent v1.0)	$0.12/s of video	—
A	Gemini Omni Flash 1.1 Google	Models	Cloud	1,132 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.1/s of video	—
A	Cosmos3-Super-Image2Video NVIDIA	Models	Local	1,275 on Artificial Analysis Video Arena: Image to Video, no audio (AA-Video-I2V-Silent v1.0)	Open weights	—
A	BACH Video Rebirth	Apps	Cloud	via BACH 1.0 Pro	Free tier	beginner
A	Adobe Firefly Adobe	Apps	Cloud	via Gemini Omni Flash 1.1	$9.99/mo	beginner
A	Flow Google	Apps	Cloud	via Gemini Omni Flash 1.1	—	beginner
A	Gemini Google	Apps	Cloud	via Gemini Omni Flash 1.1	Free tier · $4.99/mo	beginner
B	Grok Imagine Video 1.5 xAI	Models	Cloud	1,060 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.08/s of video	—
B	HappyHorse 1.1 Alibaba (ATH)	Models	Cloud	1,055 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.18/s of video	—
B	MAGI-2 Preview Sand.ai	Models	Local	1,045 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	~640 GB VRAM	—
B	Grok SpaceXSI (formerly xAI)	Apps	Cloud	via Grok Imagine Video 1.5	Free tier · $8/mo	beginner
C	PixVerse V6 PixVerse	Models	Cloud	988 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.12/s of video	—
C	Midjourney Video V1 Midjourney	Models	Cloud	1,165 on Artificial Analysis Video Arena: Image to Video, no audio (AA-Video-I2V-Silent v1.0)	—	—
C	Veo 3.1 Google	Models	Cloud	968 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.4/s of video	—
C	Gen-4.5 Runway	Models	Cloud	913 on Artificial Analysis Video Arena: Text to Video, no audio (AA-Video-T2V-Silent v2.1)	$0.12/s of video	—
C	LTX-2.5 Lightricks	Models	Local	950 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	Open weights	—
C	Vidu Q4 Preview ShengShu Technology	Models	Cloud	941 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	—	—
C	Pika 2.5 Pika Labs	Models	Cloud	838 on Artificial Analysis Video Arena: Text to Video, no audio (AA-Video-T2V-Silent v2.1)	$0.09/s of video	—
C	Ray3.2 Luma AI	Models	Cloud	1,206 on LMArena (Arena) Text-to-Video leaderboard	—	—
C	Boreal Creatify	Models	Cloud	831 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)	$0.01/s of video	—
C	HunyuanVideo 1.5 Tencent	Models	Local	1,169 on LMArena (Arena) Text-to-Video leaderboard	~14 GB VRAM	—
C	Wan 2.2 Alibaba (Wan-AI)	Models	Local	1,132 on LMArena (Arena) Text-to-Video leaderboard	~24 GB VRAM	—
C	PixVerse PixVerse	Apps	Cloud	via PixVerse V6	Free tier · $10/mo	beginner
C	Midjourney Midjourney	Apps	Cloud	via Midjourney Video V1	$10/mo	beginner
C	Canva Canva	Apps	Cloud	via Veo 3.1	Free tier · $18/mo	beginner
C	Gemini in Google Workspace Google	Apps	Cloud	via Veo 3.1	—	beginner
C	Leonardo.Ai Leonardo.Ai (Canva)	Apps	Cloud	via Veo 3.1	Free tier · $12/mo	beginner
C	SD.Next vladmandic (community project)	Apps	Local	via LTX-2.5	Free	intermediate
C	LTX Desktop Lightricks	Apps	Local	via LTX-2.5	Free	intermediate
C	Draw Things Draw Things	Apps	Local	via LTX-2.5	Free	intermediate
C	Vidu ShengShu Technology	Apps	Cloud	via Vidu Q4 Preview	Free tier	beginner
C	Amuse TensorStack	Apps	Local	via Wan 2.2	Free	beginner
–	SD.Next vladmandic (community)	Apps	Local	No public result yet	Free	intermediate
–	Civitai Civitai	Apps	Cloud	No public result yet	Free tier · $10/mo	intermediate
–	Comfy Cloud Comfy Org	Workflow platforms	Cloud	No public result yet	Free tier · $20/mo	advanced
–	daVinci-MagiHuman SII-GAIR and Sand.ai	Models	Local	No public result yet	Open weights	—
–	Dzine Dzine	Apps	Cloud	No public result yet	$8.99/mo	beginner
–	Fuser Fuser	Workflow platforms	Cloud	No public result yet	—	intermediate
–	getimg.ai getimg.ai	Apps	Cloud	No public result yet	$15/mo	beginner
–	Glif Glif	Apps	Cloud	No public result yet	Free tier	beginner
–	HuMo-17B ByteDance (Intelligent Creation Lab)	Models	Local	No public result yet	~24 GB VRAM	—
–	ImagineArt ImagineArt	Apps	Cloud	No public result yet	$5/mo	beginner
–	Kandinsky 6.0 Video Pro Kandinsky Lab	Models	Local	No public result yet	Open weights	—
–	Kling 4.0 Kuaishou (Kling AI)	Models	Cloud	No public result yet	—	—
–	Kling AI Kuaishou	Apps	Cloud	No public result yet	Free tier	beginner
–	LongCat-Video Meituan (LongCat)	Models	Local	No public result yet	Open weights	—
–	Lovart Lovart	Apps	Cloud	No public result yet	Free tier	beginner
–	Mage Mage	Apps	Cloud	No public result yet	Free tier · $10/mo	beginner
–	Meta AI Meta	Apps	Cloud	No public result yet	Free tier	beginner
–	MOVA (360p / 720p) OpenMOSS	Models	Local	No public result yet	~12 GB VRAM	—
–	NightCafe NightCafe Studio	Apps	Cloud	No public result yet	Free tier	beginner
–	P-Video-2 Pruna AI	Models	Cloud	No public result yet	$0.03/s of video	—
–	PixVerse C1 PixVerse	Models	Cloud	No public result yet	—	—
–	RunComfy RunComfy	Workflow platforms	Cloud	No public result yet	—	advanced
–	Runware Runware	Runtimes & APIs	Cloud	No public result yet	—	intermediate
–	Scenario Scenario	Apps + Workflow platforms	Cloud	No public result yet	Free tier · $15/mo	intermediate
–	SeaArt AI SeaArt	Apps	Cloud	No public result yet	—	beginner
–	SkyReels V3 Skywork AI (Kunlun Tech)	Models	Local	No public result yet	Open weights	—
–	ThinkDiffusion ThinkDiffusion	Workflow platforms	Cloud	No public result yet	—	advanced
–	WaveSpeedAI WaveSpeedAI	Runtimes & APIs	Cloud	No public result yet	—	intermediate

Benchmarks used

• Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio) by Artificial Analysis (blind human votes, Elo (pairwise human preference)) · snapshot 19 Oct 2026
  With-audio board (default title); September 2026 refresh with new prompt set and model cohort, 28 models; Kling 3.0 1080p (Pro) anchored at 1000, Kling 3.0 Omni 1080p (Pro) 1019. Kling 4.0 (launched 2026-10-15, no API yet) is not on this board as of 2026-10-19. OpenAI Sora 2, Luma Ray and Runway Gen-4.5 are not on this board. No 'last updated' date shown, as_of = date read.
• Artificial Analysis Video Arena: Text to Video, no audio (AA-Video-T2V-Silent v2.1) by Artificial Analysis (blind human votes, Elo (pairwise human preference)) · snapshot 18 Oct 2026
  Silent-video board of the September 2026 refresh (31 models, adds Runway Gen-4.5 and Pika 2.5); Kling 3.0 1080p (Pro) anchored at 1000. No Sora 2, Luma Ray, Midjourney or HunyuanVideo.
• LMArena (Arena) Text-to-Video leaderboard by LMArena (arena.ai) (blind human votes, Arena score (Bradley-Terry Elo-style rating from pairwise votes)) · snapshot 22 Sep 2026
  48 models incl. older versions; category 'overall' only in the dataset. Includes Sora 2 / Sora 2 Pro, Meta muse-video, Ray 3, HunyuanVideo 1.5, but not Kling 3.0, PixVerse V6, Vidu Q3, LTX-2.5 or HappyHorse 1.1. Read from the official dataset because arena.ai returns HTTP 403 to automated fetches.
• Artificial Analysis Video Arena: Image to Video, no audio (AA-Video-I2V-Silent v1.0) by Artificial Analysis (blind human votes, Elo (pairwise human preference)) · snapshot 18 Oct 2026
  Older v1.0 silent board with the broadest cohort (82 entries incl. Ray 3, Midjourney V1, HunyuanVideo-1.5, Wan 2.2, Runway Gen-4.5); Elo scale differs from the v1.1/v2.1 boards (Wan 2.1 14B anchored at 1000) so values are not comparable across boards. Several newest versions (Seedance 2.5, Gemini Omni Flash 1.1, MAGI-2) are missing.
• Artificial Analysis Video Arena: Image to Video (AA-Video-I2V v1.1, with audio) by Artificial Analysis (blind human votes, Elo (pairwise human preference)) · snapshot 19 Oct 2026
  With-audio image-to-video board, 23 current models; Kling 3.0 1080p (Pro) anchored at 1000, Kling 3.0 Omni 1080p (Pro) 1004; no Kling 4.0 entry as of 2026-10-19. Page notes that AA-Video-I2V v2.1 (new prompt set, broader model selection) is coming soon. Leader shown rounded (1140) on 2026-10-19; exact value carried over from the 2026-10-18 read.
• LMArena (Arena) Image-to-Video leaderboard by LMArena (arena.ai) (blind human votes, Arena score (Bradley-Terry Elo-style rating from pairwise votes)) · snapshot 22 Sep 2026
  48 models incl. older versions (overall category). Includes Kling v3 Pro, Vidu Q3 Pro, HiDream O1 Video, Ray 3, HunyuanVideo 1.5; lacks Sora 2, Meta muse-video, PixVerse V6, LTX-2.5, Runway Gen-4.5. url points to the dataset because the arena.ai image-to-video page URL could not be verified (site returns 403).