Resources

Ref2va always have some noise

#50 opened 10 minutes ago by

jiangjihua

Vllmomni minmaxh3 roadmap

#49 opened about 6 hours ago by

feizhai123

sparse-attention inference

#48 opened about 9 hours ago by

mzbac

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

👍 1

#47 opened about 12 hours ago by

cushycrux

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

➕ 1

#46 opened about 14 hours ago by

Adikkkk

Possibly a small mistake in docs/VIDEO_PROMPT_WRITING_GUIDE_base_en.md

#45 opened about 16 hours ago by

hum-ma

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

👍👀 3

#44 opened about 18 hours ago by

natalie5

LoRA Training with MiniMax H3 - Question

#43 opened about 19 hours ago by

Tomcat2048

Thanks a lot

❤️ 6

#42 opened about 21 hours ago by

DodoPapa

我是阿里的高管

😎 1

#41 opened about 21 hours ago by

wjm17173

Amazing job ! really thank you for sharing and pushing forward the video generation technology!

#40 opened about 22 hours ago by

aiclouddaily

Minimax H3 2K Upscaler?

#39 opened about 22 hours ago by

AlperKTS

Only MiniMax-H3 can do !

🔥 4

#38 opened about 23 hours ago by

sunnyboxs

非常棒的多模态能力。是否可以基于这个扩展生图能力？

#37 opened 1 day ago by

wdtd

vLLM-Omni ComfyUI integration for MiniMax-H3 (T2VA / FL2VA / Ref2VA)

🚀 1

#36 opened 1 day ago by

shunyang90

my opinion about nsfw

😎 10

#35 opened 1 day ago by

Arun63

Is there any plan to release a more lightweight model with fewer parameters?

➕ 2

#34 opened 1 day ago by

makisekurisu-jp

请问支持中文提示词吗？相关的特殊tag需要替换成中文吗？

#32 opened 1 day ago by

HappyNeNe

简直是圣人

#31 opened 1 day ago by

jyz1022

how much better is the cloud version of h3?

#29 opened 1 day ago by

BrokeAmerican

System Prompt IT2V

👀🔥 12

#28 opened 1 day ago by

rzgar

No official trainer? We open-sourced a working fine-tuning pipeline for H3

🚀 4

#27 opened 1 day ago by

ka1029

Go minimax!, fxxx seedance for ridiculous pricing.

🚀 14

#26 opened 2 days ago by

anitman

Future request

👍 2

#25 opened 2 days ago by

darkenHUB

Huge Thanks for Using a Single-Stream Transformer, Porting It to a Multi-GPU ComfyUI Setup with RayLight Is a Breeze

#24 opened 2 days ago by

komixenon

Huge thanks & support in TongFlow now

🚀🔥 1

#23 opened 2 days ago by

caotong598

感谢开源，非cfg-distilled版本以后会放出来吗

#22 opened 2 days ago by

momo5440

感谢开源

#21 opened 2 days ago by

shirubei

未来请求

#20 opened 2 days ago by

darkenHUB

where is dolly in dolly out professional camera language

#19 opened 2 days ago by

ApexArtist

Nice work, will the training recipe opensource soon?

🔥 3

#18 opened 2 days ago by

DDDPG

Please collaborate with PrismML on a Bonsai version

➕ 6

#17 opened 2 days ago by

mindplay

H3-Context-IR未来会开源吗？

#14 opened 2 days ago by

limn3022

Can this run on my 16gb i5 pc? 😭🙏

😔 2

#13 opened 2 days ago by

rithulbm

Any License Question Ask here!

👍 7

#12 opened 2 days ago by

ryanlee-dev

Would this work on 3x rtx 6000 pro?

#11 opened 2 days ago by

Quadrapole

山西人民发来贺电

#10 opened 2 days ago by

heDWE

感谢MiniMax

👍 20

#8 opened 2 days ago by

shiverwang

text encoder好大，有在线的API 吗？

#7 opened 2 days ago by

makao008

终于来了

#4 opened 2 days ago by

k3soft

Thank You

👍❤️ 4

#3 opened 2 days ago by

dns

Another common China W

🔥❤️ 11

#2 opened 2 days ago by

Nark103

上线了

❤️ 3

#1 opened 2 days ago by

pengzhi27