Resources

【招聘】 MiniMax 视频模型全球开源生态

pinned

❤️ 5

#61 opened about 1 month ago by

MiniMax-AI

Any License Question Ask here!

pinned

👍 17

#12 opened about 2 months ago by

ryanlee-dev

Guys any update for H3-Regenerate-2K?

#105 opened 3 days ago by

Meryruizk

Running on <4gb Vram.

#104 opened 11 days ago by

Jit2024

g

#103 opened 15 days ago by

jhjhjhjhjhjhjhjhjhjhjhjhjh

MiniMax Director — a ComfyUI timeline editor that covers every documented H3 feature

#102 opened 16 days ago by

imbutus

Sharing: MiniMax-H3 Character Card ComfyUI-Workflow !

❤️ 1

#101 opened 16 days ago by

sunnyboxs

Upload minimax-h3_thank-you-for-watching-genetic-testing.mp4

#100 opened 16 days ago by

hiredlawyer1

minimax

#99 opened 17 days ago by

Kiskacsa

Could you upload the 3 assets on cdn.hailuoai.com used by vllm ci testing

#98 opened 18 days ago by

cowboyBlues

Request for pre-distillation / base H3 weights

👍 8

#97 opened 18 days ago by

muchov

发现minimax的小瑕疵：眨眼问题，人物几乎不会眨眼睛

#96 opened 24 days ago by

easonchow0419

Please provide more in-depth documentation for Full-Reference Mode

👍 17

#95 opened 26 days ago by

infearia

Make new

#94 opened 28 days ago by

hkb4181

Running on <4gb Vram

🔥 2

#93 opened 28 days ago by

Jit2024

H3-Regenerate-2K open weights?

➕ 3

#92 opened 30 days ago by deleted

MiniMax H3 Ref2VA + audio recasts identity, while FL2VA + Ref2V Turbo preserves it — expected behavior?

#91 opened 30 days ago by

lorentaken

PURE MOJO STACK

#90 opened about 1 month ago by

alexnvo

蒸馏加速版本

👀 1

#89 opened about 1 month ago by

ReSerendipity

Any plans to release a non-distilled base model for training?

👀👍 6

#88 opened about 1 month ago by

AndroYD

脸崩的如此厉害，何时修复~

#87 opened about 1 month ago by

criuslv-cn

This is currently the best open‑source video model, yet the conservative algorithm of the VAE seems to create a computational bottleneck?

👍❤️ 9

#85 opened about 1 month ago by

ass5002

Getting different Accents when speaking the same language

#84 opened about 1 month ago by

Hcheyne

Ref2VA environmental audio feels too quiet in dialogue scenes

➕ 1

#82 opened about 1 month ago by

szczypen

Question about prompting on static elements

#81 opened about 1 month ago by

leonnn1

Thanks and Suggestions

#80 opened about 1 month ago by

Pvok

Omni-Rewriter Replay: observe a clip into validated H3 PE (open harness)

#79 opened about 1 month ago by

Wayne-King

Color distortion occurs when encoding and decoding images using a video VAE.

#78 opened about 1 month ago by

l13462580123

checkpoint inconsistent between original and diffusers

#77 opened about 1 month ago by

l13462580123

Weird gibberish / clipped voice at the very start of H3 videos (reference to video)

👍 4

#76 opened about 1 month ago by

jesleocizi

[FL2VA] Anime style videos have half the FPS

#74 opened about 1 month ago by

ChiNoel

Call MiniMax H3 with your existing OpenAI SDK

#72 opened about 1 month ago by

irene891107

Some question about VIDEO_PROMPT_WRITING_GUIDE_ref_en.md

#71 opened about 1 month ago by

iouzzr

Minimax H3 prompt adherence really varies depending on the resolution

#65 opened about 1 month ago by

TheBobun

How define audio ref to the Subject 1 - Ref2V

#64 opened about 1 month ago by

Gamb

第一次接触咱们这个网站，有好几个问题想要问

#63 opened about 1 month ago by

zhao7259

Update documentation to Please add how to character swap for Ref2V/Image to Video.

👍 1

#60 opened about 1 month ago by

CoolaidFun

脱领带，脱上衣，脱带搭袢的鞋等还无法正常表现

#59 opened about 1 month ago by

sumirecccp

MiniMax H3 Prompt Enhancer, powered by a fine-tuned 350M-parameter model

❤️ 3

#58 opened about 1 month ago by

geocine

Will RunPod work??

#56 opened about 1 month ago by

tonyface

Almost VR support

👍👀 5

#55 opened about 1 month ago by

Ddfgddsd

请教一下，这个模型如何生成一个可以首尾循环的视频？

#54 opened about 1 month ago by

oioitff

lora训练有什么需要注意的地方？

#53 opened about 1 month ago by

wuyuetiger

Ref2va always have some noise

❤️👀 2

#50 opened about 1 month ago by

jiangjihua

Vllmomni minmaxh3 roadmap

#49 opened about 1 month ago by

feizhai123

sparse-attention inference

#48 opened about 1 month ago by

mzbac

The actual correct system prompt for IT2V - (Took me a while)

👍 6

#47 opened about 1 month ago by

cushycrux

Where is the money, Lebowski? The "duct-tape" architecture review)))

➕ 3

#46 opened about 1 month ago by

Qozimo

Possibly a small mistake in docs/VIDEO_PROMPT_WRITING_GUIDE_base_en.md

#45 opened about 1 month ago by

hum-ma

Is this a CFG distilled model or were the weights trained without CFG from the start?

👍👀 5

#44 opened about 1 month ago by

natalie5