The toolbox
What people actually use.
38 tools for making AI video, open source and paid, ranked by stars, downloads, discussion and how often creators name them. Open any card to see exactly why it scored what it did.
-
93
#1 · Open source · GPL-3.0
ComfyUI
Node-based workflow app
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
- 137.7kGitHub stars
- 16.4kforks
- 1video mentions
- 18tutorial mentions
Why it scored 93
137,663 GitHub stars, 16,389 forks; 4 Hacker News points on tutorials naming it; named in 1 tracked videos and 18 tutorials; code pushed 0 days ago.
Evidence coverage 100% of the formula.
-
86
#2 · Paid
Seedance
Video generation
- 8video mentions
- 14tutorial mentions
Why it scored 86
4 Hacker News points on tutorials naming it; named in 8 tracked videos and 14 tutorials.
Evidence coverage 80% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
68
#3 · Paid
Runway
Video generation & editing
- 26video mentions
- 4tutorial mentions
Why it scored 68
Named in 26 tracked videos and 4 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Prices seen on the pricing page (Oct 15, 2026): $12/month. Raw text, check the page for plan details.
-
60
#4 · Paid
Sora
Video generation
- 18video mentions
- 2tutorial mentions
Why it scored 60
Named in 18 tracked videos and 2 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
55
#5 · Paid
Midjourney
Image & video generation
- 14video mentions
- 1tutorial mentions
Why it scored 55
Named in 14 tracked videos and 1 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
54
#6 · Paid
Luma Dream Machine
Video generation
- 13video mentions
- 1tutorial mentions
Why it scored 54
Named in 13 tracked videos and 1 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
52
#7 · Paid
Higgsfield
Video generation / camera FX
- 12video mentions
- 1tutorial mentions
Why it scored 52
Named in 12 tracked videos and 1 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
52
#8 · Open source · Apache-2.0
Diffusers
Library
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.
- 34.7kGitHub stars
- 7,397forks
- 0video mentions
- 0tutorial mentions
Why it scored 52
34,723 GitHub stars, 7,397 forks; named in 0 tracked videos and 0 tutorials; code pushed 0 days ago.
Evidence coverage 100% of the formula.
-
52
#9 · Paid
Veo / Flow
Video generation
- 7video mentions
- 7tutorial mentions
Why it scored 52
2 Hacker News points on tutorials naming it; named in 7 tracked videos and 7 tutorials.
Evidence coverage 80% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
51
#10 · Open source
HunyuanVideo
Video model
HunyuanVideo: A Systematic Framework For Large Video Generation Model
- 12.6kGitHub stars
- 1,353forks
- 1,887HF downloads / 30d
- 0video mentions
- 1tutorial mentions
Why it scored 51
12,622 GitHub stars, 1,353 forks; 1,887 Hugging Face downloads (30 days), 2,335 likes; named in 0 tracked videos and 1 tutorials; code pushed 7 days ago.
Evidence coverage 100% of the formula.
-
49
#11 · Open source · Apache-2.0
LTX-Video
Video model
Official repository for LTX-Video
- 11.1kGitHub stars
- 1,182forks
- 711.8kHF downloads / 30d
- 4video mentions
- 4tutorial mentions
Why it scored 49
11,106 GitHub stars, 1,182 forks; 711,765 Hugging Face downloads (30 days), 2,356 likes; 2 Hacker News points on tutorials naming it; named in 4 tracked videos and 4 tutorials; code pushed 282 days ago.
Evidence coverage 100% of the formula.
-
48
#12 · Open source · Apache-2.0
Wan 2.2
Video model
Wan: Open and Advanced Large-Scale Video Generative Models
- 17.9kGitHub stars
- 2,324forks
- 15.0kHF downloads / 30d
- 0video mentions
- 1tutorial mentions
Why it scored 48
17,945 GitHub stars, 2,324 forks; 14,984 Hugging Face downloads (30 days), 1,485 likes; named in 0 tracked videos and 1 tutorials; code pushed 24 days ago.
Evidence coverage 100% of the formula.
-
48
#13 · Paid
ElevenLabs
Voice, SFX & music
- 10video mentions
- 0tutorial mentions
Why it scored 48
Named in 10 tracked videos and 0 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
41
#14 · Paid
Hailuo (MiniMax)
Video generation
- 2video mentions
- 5tutorial mentions
Why it scored 41
Named in 2 tracked videos and 5 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
40
#15 · Open source · Apache-2.0
Wan 2.1
Video model
Wan: Open and Advanced Large-Scale Video Generative Models
- 17.2kGitHub stars
- 3,800forks
- 80.8kHF downloads / 30d
- 0video mentions
- 1tutorial mentions
Why it scored 40
17,209 GitHub stars, 3,800 forks; 80,773 Hugging Face downloads (30 days), 2,192 likes; named in 0 tracked videos and 1 tutorials; code pushed 224 days ago.
Evidence coverage 100% of the formula.
-
40
#16 · Paid
Pika
Video generation
- 0video mentions
- 4tutorial mentions
Why it scored 40
2 Hacker News points on tutorials naming it; named in 0 tracked videos and 4 tutorials.
Evidence coverage 80% of the formula.
Prices seen on the pricing page (Oct 15, 2026): $440/mo. Raw text, check the page for plan details.
-
40
#17 · Paid
Kling
Video generation
- 34video mentions
- 7tutorial mentions
Why it scored 40
1 Hacker News points on tutorials naming it; named in 34 tracked videos and 7 tutorials.
Evidence coverage 80% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
36
#18 · Open source
Wan2GP
Low-VRAM video app
A fast AI Video Generator for the GPU Poor. Supports Minimax H3, LTX-2.3/2.5, Wan 2.1/2.2, Qwen Image 1/2, Krea 2, Hunyuan Video, LTX Video and Flux.
- 10.5kGitHub stars
- 1,680forks
- 0video mentions
- 0tutorial mentions
Why it scored 36
10,522 GitHub stars, 1,680 forks; named in 0 tracked videos and 0 tutorials; code pushed 2 days ago.
Evidence coverage 100% of the formula.
-
34
#19 · Open source · Apache-2.0
Mochi 1
Video model
The best OSS video generation models, created by Genmo
- 3,751GitHub stars
- 494forks
- 8,739HF downloads / 30d
- 0video mentions
- 0tutorial mentions
Why it scored 34
3,751 GitHub stars, 494 forks; 8,739 Hugging Face downloads (30 days), 1,370 likes; named in 0 tracked videos and 0 tutorials; code pushed 8 days ago.
Evidence coverage 100% of the formula.
-
33
#20 · Open source · MPL-2.0
Coqui TTS
Voice / TTS
🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production
- 46.1kGitHub stars
- 6,166forks
- 0video mentions
- 0tutorial mentions
Why it scored 33
46,117 GitHub stars, 6,166 forks; named in 0 tracked videos and 0 tutorials; code pushed 790 days ago.
Evidence coverage 100% of the formula.
-
30
#21 · Paid
Synthesia
Avatars
- 0video mentions
- 1tutorial mentions
Why it scored 30
2 Hacker News points on tutorials naming it; named in 0 tracked videos and 1 tutorials.
Evidence coverage 80% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
30
#22 · Open source · BSD-3-Clause
Real-ESRGAN
Upscaling
Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
- 37.1kGitHub stars
- 4,496forks
- 0video mentions
- 0tutorial mentions
Why it scored 30
37,096 GitHub stars, 4,496 forks; named in 0 tracked videos and 0 tutorials; code pushed 800 days ago.
Evidence coverage 100% of the formula.
-
29
#23 · Open source · Apache-2.0
Open-Sora
Video model
Open-Sora: Democratizing Efficient Video Production for All
- 29.9kGitHub stars
- 3,094forks
- 0video mentions
- 0tutorial mentions
Why it scored 29
29,873 GitHub stars, 3,094 forks; named in 0 tracked videos and 0 tutorials; code pushed 189 days ago.
Evidence coverage 100% of the formula.
-
27
#24 · Paid
Suno
Music
- 3video mentions
- 0tutorial mentions
Why it scored 27
Named in 3 tracked videos and 0 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
26
#25 · Open source
LivePortrait
Face animation
Bring portraits to life!
- 19.2kGitHub stars
- 2,003forks
- 0video mentions
- 0tutorial mentions
Why it scored 26
19,236 GitHub stars, 2,003 forks; named in 0 tracked videos and 0 tutorials; code pushed 135 days ago.
Evidence coverage 100% of the formula.
-
24
#26 · Open source · AGPL-3.0
Video2X
Upscaling / interpolation
A machine learning-based video super resolution and frame interpolation framework. Est. Hack the Valley II, 2018.
- 22.3kGitHub stars
- 1,942forks
- 0video mentions
- 0tutorial mentions
Why it scored 24
22,264 GitHub stars, 1,942 forks; named in 0 tracked videos and 0 tutorials; code pushed 221 days ago.
Evidence coverage 100% of the formula.
-
21
#27 · Open source
Wav2Lip
Lip-sync
This repository contains the codes of "A Lip Sync Expert Is All You Need for Speech to Lip Generation In the Wild", published at ACM Multimedia 2020. For HD commercial model, please try out Sync Labs
- 13.2kGitHub stars
- 2,851forks
- 0video mentions
- 0tutorial mentions
Why it scored 21
13,242 GitHub stars, 2,851 forks; named in 0 tracked videos and 0 tutorials; code pushed 480 days ago.
Evidence coverage 100% of the formula.
-
20
#28 · Open source · Apache-2.0
FramePack
Video model / app
Lets make video diffusion practical!
- 17.3kGitHub stars
- 1,738forks
- 0video mentions
- 0tutorial mentions
Why it scored 20
17,296 GitHub stars, 1,738 forks; named in 0 tracked videos and 0 tutorials; code pushed 364 days ago.
Evidence coverage 100% of the formula.
-
19
#29 · Open source · Apache-2.0
AnimateDiff
Animation model
Official implementation of AnimateDiff.
- 12.3kGitHub stars
- 1,106forks
- 0HF downloads / 30d
- 0video mentions
- 1tutorial mentions
Why it scored 19
12,275 GitHub stars, 1,106 forks; 0 Hugging Face downloads (30 days), 1,010 likes; 2 Hacker News points on tutorials naming it; named in 0 tracked videos and 1 tutorials; code pushed 806 days ago.
Evidence coverage 100% of the formula.
-
17
#30 · Open source · Apache-2.0
CogVideoX
Video model
text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)
- 13.1kGitHub stars
- 1,373forks
- 14.8kHF downloads / 30d
- 0video mentions
- 0tutorial mentions
Why it scored 17
13,084 GitHub stars, 1,373 forks; 14,773 Hugging Face downloads (30 days), 699 likes; named in 0 tracked videos and 0 tutorials; code pushed 345 days ago.
Evidence coverage 100% of the formula.
-
16
#31 · Open source
SkyReels V2
Video model
SkyReels-V2: Infinite-length Film Generative model
- 7,681GitHub stars
- 1,689forks
- 0video mentions
- 0tutorial mentions
Why it scored 16
7,681 GitHub stars, 1,689 forks; named in 0 tracked videos and 0 tutorials; code pushed 259 days ago.
Evidence coverage 100% of the formula.
-
13
#32 · Open source · Apache-2.0
ComfyUI-WanVideoWrapper
ComfyUI nodes
- 6,732GitHub stars
- 696forks
- 0video mentions
- 0tutorial mentions
Why it scored 13
6,732 GitHub stars, 696 forks; named in 0 tracked videos and 0 tutorials; code pushed 144 days ago.
Evidence coverage 100% of the formula.
-
8
#33 · Open source · Apache-2.0
AnimateDiff-Evolved
ComfyUI nodes
Improved AnimateDiff for ComfyUI and Advanced Sampling Support
- 3,557GitHub stars
- 289forks
- 0video mentions
- 0tutorial mentions
Why it scored 8
3,557 GitHub stars, 289 forks; named in 0 tracked videos and 0 tutorials; code pushed 79 days ago.
Evidence coverage 100% of the formula.
-
1
#34 · Open source · Apache-2.0
VACE
Video editing model
[ICCV 2025] Official implementations for paper: VACE: All-in-One Video Creation and Editing
- 3,975GitHub stars
- 288forks
- 0video mentions
- 0tutorial mentions
Why it scored 1
3,975 GitHub stars, 288 forks; named in 0 tracked videos and 0 tutorials; code pushed 363 days ago.
Evidence coverage 100% of the formula.
-
0
#35 · Paid
Hedra
Talking characters
- 0video mentions
- 0tutorial mentions
Why it scored 0
Named in 0 tracked videos and 0 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
0
#36 · Paid
HeyGen
Avatars
- 0video mentions
- 0tutorial mentions
Why it scored 0
Named in 0 tracked videos and 0 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
0
#37 · Paid
Krea
Image & video generation
- 0video mentions
- 0tutorial mentions
Why it scored 0
Named in 0 tracked videos and 0 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
-
0
#38 · Paid
Topaz Video AI
Upscaling
- 0video mentions
- 0tutorial mentions
Why it scored 0
Named in 0 tracked videos and 0 tutorials; no public community-rating source we can use within ToS, so score rests on mentions only.
Evidence coverage 35% of the formula.
Pricing not captured (page is script-rendered or not checked). See the official site.
Rising on GitHub
Popular repositories tagged text-to-video that aren't in the directory yet. Candidates for review, not ranked.
- calesthio/OpenMontage★ 68.5kWorld's first open-source, agentic video production system. 12 production pipelines, 100+ tools, 700+ agent sk
- Anil-matcha/Open-Generative-AI★ 30.4kUnrestricted Open-source alternative to AI video platforms — Free AI image & video generation studio with 600+
- hypit-ai/hypit★ 22.4kClone any viral video with AI agents. Not just a script, the whole workflow: swap the face, the words, the B-r
- HBAI-Ltd/Toonflow-app★ 17.8k短剧、漫剧制作工具 Toonflow：开源 AI 视频创作平台，集成剧本创作、智能分镜、角色与场景资产管理、图像生成和视频生成。融合无限画布、AI Agent 与可视化工作流，支持本地部署、自选模型、跨平台客户端、MCP
- NVlabs/Sana★ 9,341SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer
- lucidrains/imagen-pytorch★ 8,431Implementation of Imagen, Google's Text-to-Image Neural Network, in Pytorch
- promptslab/Awesome-Prompt-Engineering★ 6,373This repository contains a hand-curated resources for Prompt Engineering with a focus on Generative Pre-traine
- dramaclaw/dramaclaw★ 6,299A general-purpose AIGC video engine: script to finished film in one pipeline — dramas, ads, product videos, ot
- ArcReel/ArcReel★ 5,514AI Agent 驱动的开源可自部署视频工作台：将小说与剧本转为角色、场景、道具资产、分镜、视频和剪映草稿，支持跨镜头一致性、多供应商与费用追踪 | Self-hosted AI video workspace for
- AILab-CVC/VideoCrafter★ 5,101VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models
- Tencent-Hunyuan/HunyuanVideo-1.5★ 4,570HunyuanVideo-1.5: A leading lightweight video generation model
- YangLing0818/Diffusion-Models-Papers-Survey-Taxonomy★ 3,365Diffusion model papers, survey, and taxonomy
How tools are scored
score = 100 × ( 0.45·Community + 0.35·Mentions + 0.2·Activity ) ÷ (weights with data) × (0.6 + 0.4·coverage)
- Community
- Average of each available signal, log-scaled then min-max normalized across all tools: GitHub stars, forks, Hugging Face 30-day downloads and likes, and Hacker News points on stories that name the tool.
- Mentions
- How often the tool is named in titles and descriptions of the 280 videos on the Watch page plus the tutorials on Learn, log-scaled and normalized.
- Activity
- Open source only: e−days since last push / 90.
- Coverage
- The share of formula weight we actually have data for. Missing signals are left blank, not guessed; a tool scored on mentions alone is discounted rather than inflated.
Paid tools have no public rating source we can use within terms of service (Product Hunt and G2 block or forbid automated access), so they lean on mentions. Comment sentiment is not scored yet.