Star 历史趋势
数据来源: GitHub API · 生成自 Stargazers.cn
README.md

MiniMax H3

MiniMax H3 Integrations

A community-maintained index of models, tools and projects for MiniMax H3 and MiniMax-H3-Image.

English | 简体中文

Contents

MiniMax H3

MiniMax H3 is MiniMax's open-weight model for video with native stereo audio: it reads text, images, video and audio as context and generates clips of 4 to 15 seconds, up to 2K. It comes in two variants: FL2VA works from text and optional first and last frames, and Ref2VA takes up to 9 images, 3 videos and 3 audio clips as references.

Official

MiniMax-H3Official repository: reference implementation and prompt guides.
MiniMaxAI/MiniMax-H3Original weights for FL2VA (first and last frame) and Ref2VA (reference), with text encoder, video VAE and audio VAE.
Comfy-Org/MiniMax-H3ComfyUI single-file weights, including AdaLN-pruned BF16, FP8 and INT8 builds.
ComfyUI tutorialOfficial ComfyUI guide for H3, from the day-0 announcement.
ComfyUI workflow templatesOfficial T2V, I2V and R2V templates that ship with ComfyUI.
Video prompt writing guidesOfficial guides for base (FL2VA) and reference (Ref2VA) prompts.
MiniMax APIOfficial video generation API; for mainland China, use platform.minimaxi.com.
Hailuo AIUse H3 in the browser or the desktop app; for mainland China, use hailuoai.com and hub.minimaxi.com.
ComfyUI API nodesBuilt-in nodes for text, first and last frame and reference to video that run H3 and H3 Max through your Comfy account.
MiniMax CLIOfficial command-line tool for the MiniMax API, including H3 video generation.

Quantizations

unsloth/MiniMax-H3-GGUFPruned FL2VA and Ref2VA GGUF from Q2_K (6.26 GiB) to Q8_0, plus Qwen3-VL text-encoder GGUF.
DmitryDB/MiniMax-H3-ComfyUI-QuantsINT8 ConvRot and NVFP4 builds of FL2VA and Ref2VA from 10.86 to 21.91 GiB, plus VAE files.
DmitryDB/MiniMax-H3-DynTime-sQKVINT8 DT-sQKV builds that need the repository's ComfyUI runtime patch.
rockerBOO/minimax-h3-nvfp4-convrotNVFP4 FL2VA and Ref2VA for Blackwell GPUs, recommended with INT8 ConvRot; experimental INT4 ConvRot FL2VA for other GPUs.
Abiray/Minimax-H3-nvfp4-INT4-INT8-ConvrotPruned NVFP4 (11.67 GiB), INT8 ConvRot and mixed INT4/INT8 ConvRot builds.
tsolful/Minimax_H3_INT4MixedConvRotPruned mixed INT4/INT8 ConvRot builds of FL2VA and Ref2VA.
Merserk/MiniMax-H3-INT4-ConvRotINT4 ConvRot, 16.67 GiB full and 10.56 GiB pruned.
AX1Y2JP/MiniMax-H3-W4A8-ConvRotPruned W4A8 ConvRot DiT (11.68 GiB) and a W4A8 text encoder.
Kijai/MiniMax-H3-experimentalExperimental W4A8 mixed DiT builds and an INT8 ConvRot video VAE.
Winnougan/MiniMax-H3-INT4_Convrot_ComfyUIPruned W4A8 ConvRot DiT with a matching quantized text encoder.
rzgar/minimax_h3_fl2va_fp8_e4m3fnFP8 E4M3FN, MXFP8, and FP8 with FP16 attention builds of FL2VA.
abakanai/Minimax_h3_hybridHybrid NVFP4 Ref2VA builds for Blackwell GPUs, 14.03 and 16.38 GiB.
DiffSynth-Studio/MiniMax-H3-NF4NF4 DiT, text encoder and VAEs for DiffSynth-Studio, from 8 GB of VRAM.
WaveCut/MiniMax-H3-OrbitQuant-W4A4W4A4 build with a quantized text encoder; needs the ComfyUI-OrbitQuant node.
Abiray/MiniMax-H3-GGUFFull FL2VA and Ref2VA GGUF from Q3_K_S to Q8_0, with text-encoder quants and a workflow.
Abiray/MiniMax-H3-Pruned-GGUFPruned FL2VA GGUF, Q3_K_M and Q8_0.
realrebelai/MiniMax-H3_GGUFsQ3_K_M and Q4_K_M GGUF for FL2VA and Ref2VA, plus Q2_K and Q4_K_M text encoders; the Q2_K one is 7.91 GiB.
vantagewithai/MiniMax-H3-comfyUI-GGUFFull FL2VA and Ref2VA GGUF from Q3_K_M to Q8_0 for ComfyUI.
MarxistLeninist/MiniMax-H3-FL2VA-Pruned-IQ1-GGUFIQ1_S (3.78 GiB) and IQ1_M builds, the smallest published DiT.
molbal/MiniMax-H3-GGUFPruned Ref2VA GGUF, including hand-tuned mixed layouts.
leejet/MiniMax-H3-GGUFPruned FL2VA GGUF, Q2_K and Q4_K_M.
joeygambino/MiniMax-H3-curve-GGUFPruned FL2VA and Ref2VA GGUF in Q4_0, Q5_1 and Q8_0; needs the H3-Multishot patch for ComfyUI-GGUF.
jfar-z/MiniMax-H3-FL2VA-Ref2VA-Hybrid-NVFP4NVFP4 builds of the smhfacct FL2VA and Ref2VA hybrid in four block ranges, 11.67 GiB each, for Blackwell GPUs.
lilcheaty/MiniMax-H3-NVFP4Pruned NVFP4 FL2VA and Ref2VA at 11.67 GiB each, plus full and mixed Ref2VA builds, for Blackwell GPUs.
DeepBeepMeep/MiniMax-H3Weights WanGP downloads for the H3 family: FL2VA, Ref2VA, FastH3 Omni, HyperFlow Flash, Fun ControlNet and H3-Image, in BF16 and INT8.

Components

alibaba-pai/MiniMax-H3-Fun-Controlnet-UnionOne control branch for Canny, depth, HED, MLSD and pose, plus video inpainting.
lihaoyun6/MiniMax-H3-Ref-Patch148 MiB patch that lets FL2VA behave more like Ref2VA; needs its ComfyUI node.
Kijai/MiniMax-H3-TAE9 MiB tiny autoencoder for previews in ComfyUI-KJNodes.
Mamad8/MiniMax-H3-Image-VAEExperimental single-frame VAE for still images from the video model; keep the original VAE for video.
LBH-123-AI/Minimax_h3_latent_UpscalerLearned latent upscaler weights, 1.0 to 4.0x, for its ComfyUI node.
Open-H3-Regenerate-2KNVIDIA's three-step refiner that takes a low-resolution H3 draft to 2560x1440, with code.
ZawShiShawn/Gestura-MiniMax-H3-VAE-LiteRTH3 VAE in LiteRT format for Gestura, an Android app for on-device video.
speach1sdef178/MiniMax-H3-X2-Detail-VAEVideo VAE that decodes at twice the resolution; needs the fast VAE node from ComfyUI-MiniMaxH3_LatentUpscaler.
Efficient-Large-Model/Open-H3-Regenerate-2K-ICIn-context LoRA version of NVIDIA's refiner that redraws a 768p draft at 1080p, 2K or 4K through Ref2VA.
joeygambino/MiniMax-H3-encoder-GGUFGGUF builds of the Qwen3-VL-32B text encoder for ComfyUI.

Acceleration and distillation

Minimax-H3-TurboLightX2V's 4- and 8-step distillation LoRAs, with weights on Hugging Face.
larryvrh/MiniMax-H3-Turbo-LoraCommunity 4-step Turbo re-bakes; v4_step600_ema is the usual default.
drbaph/MiniMax-H3-Turbo-Lora-ComfyUITurbo LoRAs converted for the pruned ComfyUI checkpoints, including rank-21 resizes.
Abiray/MiniMax-H3-Turbo-Lora-Pruned-ComfyUITurbo LoRAs for the pruned ComfyUI checkpoints.
Kijai/MiniMax-H3_comfyLightX2V Turbo LoRA in ComfyUI format, full and rank-21.
InstantX/MiniMax-H3-Turbo-Lora-Diffusers4-step Turbo LoRA in diffusers format.
joyfox/MiniMax-H3-Turbo4-step FL2VA LoRA.
Tutu 20-to-8 NFE LoRAAudio-video LoRA that cuts 20 evaluations to 8.
t8star/minimax-h3-4step-turbo-loras-comfyui-expExperimental INT8 ConvRot Turbo LoRAs for ComfyUI.
infosave/MiniMax-H3-Turbo-cmfTurbo checkpoints in CMF format, Q4TP and Q2TP.
alibaba-pai/MiniMax-H3-Acc-LoRAsPDD 8-step acceleration LoRAs for FL2VA and Ref2VA, loaded natively by current ComfyUI.
aptech0081/MiniMax-H3-PDD-ComfyUIPDD LoRAs converted for ComfyUI, plus INT8 Ref2VA checkpoints with PDD baked in.
t8star/Minimax-H3-Super-Acceleration-ComfyPDD LoRAs repacked for the T8 audio nodes.
1ronman1993/MiniMax-H3-SVDQuant-fp4-pdd8SVDQuant FP4 build with the PDD 8-step LoRA folded in; an INT4 build is also available.
FastH3 4-step PreviewFastVideo's data-free 4-step checkpoint with sparse attention; also as a LoRA.
FastH3 8-Step V28-step text-to-audio-video checkpoint with 80% sparse attention.
FastH3 Trim NVFP4Experimental 8-step model with 42 of 50 blocks, 11.1 GiB in NVFP4 for Blackwell and DGX Spark.
FastH3-Omni 8-Step8-step reference-conditioned video and audio from image, video and audio references.
FastVideo/FastVideo-FastH3-ComfyFastH3 8-Step V2 repackaged for ComfyUI.
realrebelai/FastH3_GGUFsGGUF conversions of FastH3 4-step.
Pika H3 FlexPika's six-step FL2VA and Ref2VA adapters for native 480p and 768p video, with CLI, diffusers and ComfyUI examples.
PTDEPFL's four- and six-step trajectory distillation, with weights and ComfyUI integration.
DMADFour- and eight-step text-to-audio-video LoRAs with weights, workflows and training code.
VDNVideo Delta Net: hybrid linear and softmax attention for H3, 2.6 to 2.9x faster per step, with weights and training code.
drbaph/vdn-minimax-h3-int8-convrot-comfyuiINT8 ConvRot VDN in ComfyUI layout.
t8star/Vdn-Minimax-H3-ComfyReady-to-use VDN model bundle for the T8 nodes.
RAVEN Streaming LoRA4-step preview that turns H3 into a causal streaming generator, run through the RAVEN code.
MATLOWAI/minimax-h3-fused-turbo-int8-convrotOne 21 GB INT8 ConvRot file with the Turbo and motion LoRAs folded in, for 4-step text-to-video, image-to-video and Ref2VA in ComfyUI.
HyperFlow H3 FlashVideo Rebirth's 13.7B compressed H3 for video with stereo audio in 4 to 8 steps, INT8 and GGUF weights; also an 8-step LoRA for full H3.
TaoMate-H3Three-step streaming generation for H3, with weights.
TurboDiffusionW8A8 kernels, sampler and multi-GPU runtime for H3 FL2VA, with TurboH3-FL2VA-I2VA weights.
LongLiveNVIDIA's LongLive-Plug distillation, with few-step and CFG adapters for H3.
FastGenNVIDIA's distillation framework, with H3 among its video-with-audio models.
cache-ditCaching and parallel inference for diffusers, covering H3 T2VA, FL2VA and Ref2VA.
Veda-Sparse/Minimax-H3-T2VA-Veda-8NFE-600Step-PreviewLearned sparse attention keeping about 10% of tiles in every mode, plus a tuned R2VA predictor; runs in ComfyUI or Miowtion.
lightx2v/Minimax-h3-Turbo-SLA4-step FL2VA LoRA with 85% sparse-linear attention, about 2.5x faster on an RTX 5090 in LightX2V, with a ComfyUI build.
taobao-content-ai/MiniMax-H3-PTD-LoRAsPer-step PTD LoRAs that run Ref2VA in 4 or 8 steps, with diffusers code and ComfyUI nodes that need ComfyUI-PTD and the unpruned Ref2VA model.
pdmd2026/pdmd_4NFE_loraRank-128 LoRA for 4-step generation, distilled with projected distribution matching, in diffusers key format.
ChrisColeTech/minimax-h3-turbo-GGUFPruned FL2VA and Ref2VA GGUF with the LightX2V Turbo LoRA baked in, plus 4B and 8B text encoders for ComfyUI-ClipProj.

LoRAs and fine-tunes

minimax-h3-turnaroundFive rotated views of one subject from a single reference image.
Realism - Peoplefal's LoRA for natural-looking people in everyday scenes.
Looping Sketch AnimeHand-drawn 2D anime look, made to loop.
Krea2 StyleExperimental style LoRA toward Krea 2's look, for text-to-video only.
smhfacct/Minimax-H3-fl2va-ref2va-hybrid-modelsFL2VA base with Ref2VA AdaLN blocks merged in, in four block ranges.
t8star/minimax_h3_ref2va_patchin_hf102Experimental Ref2VA weight modification that boosts one high-frequency patch by 2%.
Viggle RemakeTwo Viggle LoRAs that recast people or change objects, backgrounds and weather; Viggle Animate recasts from one repainted frame.
P-H3-ImagePrunaAI's fine-tune of the video model for text-to-image and image editing.
Alissonerdx/Minimax-H3-ComfyUIExperimental LoRAs for motion transfer, character and head swap, style transfer, VFX edits and live wallpapers.
360 OrbitFL2VA LoRA that turns one photo into a frozen-time 360° orbit.
Minimax-h3 SingularityFine-tuned merge of the FL2VA and Ref2VA checkpoints for text-, image-, reference- and video-to-video in ComfyUI; also a pruned build.
prithivMLmods/MiniMax-H3-Facial-Realism-CloseUpExperimental rank-16 LoRA for close-up faces with subtle expressions and micro-movements; prompts start with Facial Realism.
LiconStudio/Minimax-H3-VBVRImage-to-video LoRA trained on rebuilt VBVR data that reduces duplicated objects and characters passing through objects.
alvdansen/h3-keyframe-animationThree LoRAs that hold a hand-drawn animation cadence from one or two drawings; free for individuals and small businesses, gated.
XGENlabs/XGEN-JINGFour-step first-person model on Ref2VA that makes video and audio from keyboard camera moves, reference images and text.
SolarWMWorld model whose H3 release streams camera-controlled video from a first frame and a prompt, with training code and the 4-step SolarRenderer.
Wushu Action LoRAMartial-arts action LoRA trained on 924 wushu clips; trigger word wushu_action.

Runtimes

ComfyUINative support since day 0, with INT8 in mainline.
DiffSynth-StudioMiniMaxH3Pipeline with NF4 inference from 8 GB of VRAM.
LightX2VInference with parallelism, quantized DiT and feature caching; home of the Turbo LoRAs.
Sana Sol-EngineNVIDIA's per-GPU optimized runtimes, 3.55 to 4.52x faster than baseline, and the 4-step Sol-H3 runtime for up to 8 GPUs.
FastVideoSparse-attention and step-distillation framework behind FastH3.
VideoX-FunTraining and inference code behind the Fun ControlNet-Union and the PDD LoRAs.
h3.cNative Metal engine for Apple silicon: text-to-video, first and last frame, and Ref2VA.
Draw ThingsSwift and Metal engine behind the Draw Things app: FL2VA, Ref2VA and FastH3 Omni on Apple silicon.
mlx-serveApple silicon app and server with FastH3 Omni reference video in 8 steps.
RaylightMulti-GPU ComfyUI runtime that splits sequences across GPUs.
UnslothLists MiniMax-H3 among the models it can run and train.
h3-flashInference scripts and a web demo for 8x B200 with sequence parallelism and fused kernels.
TorchTPU + DiffusersH3 on Cloud TPU v6e-4 with native PyTorch, torch.compile and tensor parallelism.
stable-diffusion.cppC/C++ inference engine that runs H3 and its GGUF builds.
LocalAILocal OpenAI-compatible server; version 4.9.0 added H3 video with audio.
fastllmNative H3 text-to-audio-video with an HTTP demo.
audio.cppC++ engine that runs H3 text-to-audio-video from Q4_K GGUF, optionally with an INT8 ConvRot DiT.
1Cat-vLLMvLLM fork for V100 with native H3 audio-video generation.
gufoInference engine for AMD Strix Halo that runs H3 video with sound.
vllm.cppPython-free C++ engine that serves H3 from Q4_K_M GGUF through /v1/videos, measured at 34.6 s per step on one Jetson Thor.

Apps and studios

open-videoRuns H3 on your own GPU with install, pull and run, plus a skill that lets coding agents make video.
FreeVideoLocal ComfyUI workspace built on VDN-H3, with an 8 GB VRAM path.
OpenjutsuCharacter and object swaps in video with SAM 3.1 tracking.
H3 MotionEditor for keyframed subject motion and masked region regeneration.
MiniMax-H3 StudioSelf-hosted web and API studio with a persistent queue, including FastH3 and Pika H3 Flex.
TaleMondo StudioNoncommercial Mac workspace for film production with local H3.
Tiiny StudioDesktop app for Tiiny Pocket hardware with H3-Turbo text- and image-to-video.
LF StudioWeb studio for NVIDIA's Sol-H3 pipeline on one DGX Spark or one RTX 5090; formerly sol-h3-studio.
Omnichar StudioCharacter studio that keeps one face, body, outfit and voice across shots; generates with H3 and trains H3 LoRAs locally. Formerly Inline Studio.
WanGPVideo generator for low-VRAM GPUs, with H3 among its models.
SwarmUIWeb UI with H3 among its video models.
PallaidiumGenerative movie studio inside Blender's video editor, with H3 and the Open-H3 Regenerate upscaler.
PhospheneNative Mac app that runs H3 and LTX-2.5 offline on Apple silicon; paid Pro version optional.
freefieldReady-made H3 apps on Modal: video from text or a photo, extending a clip, and new camera angles.
nightmare-mobileAndroid node-graph app for Snapdragon phones that makes H3 video with sound; non-commercial license.
infinite-creationAgent app that turns a novel or script into a video of any length with local ComfyUI H3; needs 12 GB or more of VRAM.
AmuseDesktop app built on .NET for local image, video and audio generation that lists H3 among its video pipelines, on CUDA or Vulkan.
ltx-video-macNative macOS app that runs H3 on Apple silicon through h3.c, in BF16, INT8 or a 6-step Turbo build, with a local REST API.
CodeVideoModelHas a language model build a Three.js layout video, then turns it into realistic video with local H3 Ref2VA.
GameFactory-3AGame-asset framework whose CG-video stage renders cutscenes and trailers with local H3 at 720p.
MiniMax H3 Video StudioSelf-hosted studio and CLI on ComfyUI for H3 text, image, reference and video-to-video, and segment-by-segment long videos.
ShinsekaiAI RPG maker that can animate character sprites with H3 image-to-video on local ComfyUI.
OpenShortsLocal-first short-video pipeline whose AI drama mode generates clips with H3 on your machine through stable-diffusion.cpp, or in the cloud.

API integrations

Hermes AgentSelf-improving agent whose fal video plugin runs H3, H3 Max and H3 Max Turbo, mapping usual resolutions to their tiers.
n8nWorkflow automation tool whose MiniMax node can generate H3 video through the MiniMax API; fair-code license.
MoneyPrinterTurboTurns a topic or keyword into a short video and can generate the footage with H3 through Metaso's API.
OpenMontageAgentic video production system with an H3 skill and tools that call the MiniMax API, fal or local ComfyUI.
LiteLLMPython SDK and AI gateway that can create H3 video tasks on the MiniMax API through its OpenAI-style video calls.
New APISelf-hosted AI gateway whose Hailuo task plugin sends H3 multimodal jobs to MiniMax's v2 video API.
PuterSelf-hostable web desktop whose MiniMax video provider uses H3 by default.
OpenMAICMulti-agent classroom whose video adapter runs H3 and H3 Max on MiniMax's v2 API, directly or via TokenDance.
AgnoAgent framework whose MiniMax tool generates H3 video from a text prompt through the MiniMax API.
ArtCraftDesktop app that stages scenes in 2D and 3D, then generates video with H3 through its own cloud or fal.
QwenPawPersonal assistant whose Creator plugin makes video with H3 or H3 Max on the MiniMax API, or with self-hosted H3 on SGLang.
MagpieMenu-bar app that sets up models for coding agents; with a MiniMax API key it also makes videos with H3.
SimAgent workflow workspace whose video tool runs H3 and its Highspeed, Lite and Max versions on the MiniMax API.
AI SDKVercel's TypeScript toolkit whose MiniMax provider makes H3 and H3 Max video and checks each mode's input limits.
MaxKBEnterprise agent platform whose MiniMax provider adds H3 and H3 Max text-to-video and image-to-video models.
ToonflowShort-drama studio from script and storyboard to video, calling H3 through the Metaso, CompShare or APIMart API.
Huobao DramaGenerates a short drama from one sentence to finished video, with an H3 adapter for the MiniMax API; non-commercial license.
CorsairApp integration platform whose MiniMax plugin lets agents generate H3 and H3 Max video through the MiniMax API.
waoowaooFilm and video production platform that calls H3 and H3 Max through OpenRouter; Elastic License 2.0.
BifrostAI gateway that runs H3 and H3 Max as async tasks on MiniMax's v2 API, fitting size and seconds to each model's range.
RikkaHubAndroid chat app for many LLM providers that can also generate H3 video through the MiniMax API.
xiaobeiMarketing agent for small businesses whose video skill makes clips with H3 through the MiniMax API; MIT with extra terms.
JiuwenSwarmAI agent whose video tools send H3 and H3 Max jobs to the MiniMax API.
iPolloWorkAgent workbench with H3 video and H3 digital-human tools run as RunningHub workflows; personal non-commercial license.
ArcReelAgent workspace that turns novels and scripts into shots and video, with a built-in H3 endpoint for the MiniMax API.
MonidMarketplace of tool APIs for agents, with connectors for H3, H3 Max and H3 Max Turbo on MiniMax's v2 API.
RavenMulti-agent system whose MiniMax media tools create H3 and H3 Max videos through the v2 API.
PollinationsOpen-source generative AI platform whose API serves H3, H3 Max and H3 Max Turbo video relayed from fal.
MonkeyCodeEnterprise AI office workbench whose video service runs H3 text, frame and reference modes on the MiniMax API.
YouTube Automation AgentRuns a YouTube channel end to end and can render clips with H3 on the MiniMax API under a paid-seconds cap.
RubyLLMRuby AI framework whose MiniMax video protocol lets its animate call create H3 videos.
AgentFieldFramework for running agents as services whose MiniMax media provider uses H3 as its default video model.
Agency OrchestratorMulti-agent workflow tool whose video steps can render H3 clips through the Metaso or APIMart API.
LocalMiniDramaShort-drama tool that keeps data on your machine and renders shots with H3 on the MiniMax API or through relays.
BeefTVCanvas video workspace with a live H3 Max Director view on fal and an H3 plugin for MiniMax's v2 API; source-available.
OhMyGameAI game studio app whose video tools offer H3 and H3 Max through the MiniMax API.
open-ai-canvasWorkspace for AI film with H3 prompt cards and H3 video through MiniMax, Alibaba Bailian or Metaso.
ai-fusion-videoJava agent platform for short dramas and comics, with H3 and H3 Max presets for the MiniMax API.
Row-BotLocal-first assistant whose video tool runs H3 and its Highspeed, Lite and Max versions on the MiniMax API.
AICONCanvas workflow agent for video creation whose default video model is H3, called through the project's own API gateway.
infinite-canvasAI creation canvas that makes H3 video over a MiniMax-protocol channel, with built-in H3 prompt-writing skills.
ClipForgeTurns one product photo into a short sales video, with H3 and H3 Max available through Atlas Cloud.
StoryGen-AtelierStoryboard tool where Gemini writes shots and frames, with H3 on the MiniMax API as an alternative to Veo for clips.
RevericDesktop studio for novels and short dramas that generates video with H3 through the MiniMax or Metaso API.
TapCanvasAgent-native board for AI film and comics whose bundled gateway speaks MiniMax's H3 v2 video protocol.
wind-comicMulti-agent pipeline that turns one line of text into a short drama, rendering shots with H3 on the MiniMax API.
ReqLLMElixir LLM library whose generate_video calls H3 on the MiniMax API and can upload local reference files.
KivioScreen-level AI assistant for macOS and Windows that can make H3 videos through MiniMax's v2 API.
OpenTryOnVirtual try-on and fashion toolkit with H3 and H3 Max through the MiniMax API, or H3 on a local GPU; non-commercial.
Open Storyboard CanvasAI shot-planning board with camera controls whose video nodes can use a preset MiniMax H3 provider.
LoomicChat-driven design agent on an infinite canvas that makes H3 video from text or frames through Metaso's API.
KinemaAgent for AI comics, films and commentary videos, with a minimax-h3 provider on MiniMax's v2 API.
BrewReelTurns a product brief into a vertical promo video; its talking-head mode adds AI B-roll made with H3 on the MiniMax API.
AI Video Production EditorElectron app for AI filmmaking whose node pipelines can run H3 text- and image-to-video on fal.
MoraLocal-first promo video workbench that can render clips with H3 through Atlas Cloud.
Polox AIGeneration platform built on DeepSeek Harness, with H3 text, image and reference to video through WaveSpeed.
ChibiSelf-hosted Telegram bot whose video generation can use H3 and H3 Max through the MiniMax API.
WaveSpeed MCP ServerMCP server whose H3 family generates, edits and extends video on WaveSpeed's API.
Open Media GatewaySelf-hosted service whose MiniMax provider covers H3 and its Highspeed, Lite and Max versions on the official API.
KunFlixPlatform for film and TV ad creation whose video service calls H3 through MiniMax's v2 API; Apache-2.0 with extra terms.
hf-studioSelf-hosted media studio that runs H3 video through Higgsfield and H3-Image through KIE.
MagivizAI video creation engine whose video models include H3 through KIE's API.
OpenFilmVideo agent and editor for coding agents that can replace the people in a clip with H3 Max Recast on fal.
CozyClayBrowser previs tool: block a scene and camera, then render shots with fal's H3 Max Turbo or your own ComfyUI H3 workflow.
OpenStoryVideo sequence platform whose motion step offers H3 Max text, image and reference to video on fal.
video-to-game31 games, each rebuilt from a 10 to 15 second H3 Max clip made on fal, with the asset pipeline.
anymoveMotion capture for any creature from two-camera H3 Max clips on fal, with a skill that trains an H3 LoRA there.
prompt2rigType a motion and a rigged character performs it, with H3 Max Turbo on fal as the fast video option.
falcraftMinecraft mod for fal-generated builds, including a wall TV whose live stream is rendered by H3 Max Director on fal.
varg SDKTypeScript video SDK with H3 Max text, image and reference to video models on fal; its JSX API is now deprecated.
Banana StudioSelf-hosted media studio with H3 Max text, image and reference to video, extend, lip sync and recast on fal.
Henji AIDesktop app that brings many AI providers together, with H3 Max through fal or APIMart.
pokemonliveLocal Pokémon battle prototype whose cinematics come from H3 Max Turbo and H3 Max reference-to-video on fal.
live-classroomLive TV channel that teaches any topic, with each lesson beat rendered just in time by H3 Max Turbo on fal.
H3 Max Realtime EducationScience tutor you can interrupt by voice, with teaching video generated continuously by H3 Max on fal.
SketchMakerSelf-hostable creative app whose animate, story and lip-sync features use H3 Max on fal by default.
H3 Max BlenderBlender neural rendering demo: H3 Max on fal restyles a gray ship in four looks as its geometry changes.
SPITEPre-production canvas for filmmakers whose video nodes can use H3 Max through fal with your own key.
Interdimensional GameChoose-your-own-adventure game with real-time video from H3 Max on fal.
UNREELAI streaming app where a language model writes each episode and H3 Max Turbo on fal renders it shot by shot.
StoryCastTurns a topic into a story video, with shots and lip-synced lines from H3 Max on fal.
OpenFlipbookClickable picture book where every page is AI-drawn, with optional page animation from H3 Max or H3 Max Turbo on fal.

ComfyUI nodes

comfyui-minimax-h3-audio-T862 nodes for audio, stills, conditioning, long video and speech.
ComfyUI_MiniMaxH3_Director (huangserva)Storyboard nodes with five importable templates: t2v, fl2v, r2v, v2v and rv2v.
ComfyUI_MiniMaxH3_Director (AIMixer)The original Director nodes.
ComfyUI-MiniMaxH3-DirectorTimeline editor: media on tracks, one prompt per shot, live preview and retakes.
ComfyUI-MiniMaxDirectorTimeline of shots, camera moves and audio cues compiled into the structured prompt from the official guides.
ComfyUI-MiniMaxH3-EasyOne compact workflow for T2V, I2V, first and last frame, and reference video.
ComfyUI-FeiHou-Easy-H3Fork of MiniMaxH3-Easy with built-in slots for 9 images, 3 videos and 3 audio references.
ComfyUI-H3-Motion-ContextChains clips while keeping motion and sound continuous across them.
ComfyUI-PainterNodesMiniMaxRefToVideo2 node with the official reference and dialogue format.
ComfyUI-H3-FunControlLoads the Fun ControlNet-Union with its preprocessors and inpaint path.
ComfyUI-MiniMaxH3_Ref-PatchApplies the Ref Patch to FL2VA.
VPackCompresses Ref2VA reference-video conditioning to cut reference tokens.
TARA-H3Experimental temporal alignment of Ref2VA video-reference attention, with a technical report.
ComfyUI-DialsTraining-free steering vectors built from prompt pairs, for motion and style.
LanPaintTraining-free video and audio inpainting.
ComfyUI-MiniMaxH3_LatentUpscalerLatent upscaler for H3's audio-video latents, for two-pass sampling.
Comfyui_Minimax_h3_latent_UpscalerLearned 1.0 to 4.0x latent upscaler that skips the VAE round-trip.
ComfyUI-H3-ResCascadeHires-fix nodes that refine about 1 MP renders to 1080p or 1440p, keeping the audio.
ComfyUI Open-H3 RegenerateNodes and workflows for NVIDIA's Open-H3-Regenerate-2K refiner.
TesseraTiled, chunked video refinement with streamed VAE encode and decode.
ComfyUI-H3VAE_TRTTensorRT video VAE, up to 1.7x faster decode.
ComfyUI-Spectrum-MiniMax-H3Skips selected transformer steps by forecasting features; output is approximate.
ComfyUI-SolAttn_tritonTraining-free Sol-Attn sparse attention, tested with H3; deprecated now that ComfyUI has it built in.
ComfyUI-VDN-H3Runs the VDN adapters in ComfyUI; a 24 GB fork offloads more.
ComfyUI-Ref2VA-VSAVideo sparse attention for Ref2VA, about 72 s per 5 s clip on an RTX 4090.
ComfyUI-MiniMax-H3-PDD-AccLoads the PDD LoRA with its 32 output heads on older ComfyUI.
ComfyUI-Sol-H3ComfyUI front end for the Sol-H3 runtime (GPL-3.0).
Spark-H3Adaptive block-sparse attention with diffusers and ComfyUI packages.
ComfyUI-OrbitQuantLoads the OrbitQuant W4A4 build.
ComfyUI-FastH3-Lora-ConverterConverts the FastH3 LoRA to ComfyUI key order.
ComfyUI-INT8-FastLargely superseded by native INT8; its convert_comfy_quant.py converts older INT8 files.
ComfyUI-MiniMaxH3-TimelineDirectorEditable timeline of reference media for Ref2VA, with an all-in-one director workflow.
ComfyUI-MiniMaxH3ModRefMod adapters that compress reference images and videos into small latents without training; still under construction.
ComfyUI-H3-FaceRefineTracks small faces frame by frame, refines them with H3 and stitches them back.
comfyui-AICG3DShort-drama workbench that splits a script into segments for videos of any length.
H3-Character-Sheet-DesignerCharacter reference sheets with up to 13 views from one character image.
ComfyUI_MiniMax_H3_ExtenderChains clips into long video while keeping motion, picture and audio continuous.
ComfyUI-Majoor-H3-GuideMasterPlaces images and audio on the H3 timeline as guides.
ComfyUI-Viggle-Animate-H3Nodes for Viggle Animate, which replaces a character in a driving video with a reference.
ComfyUI-MiniMax-H3-Image-StudioStill images and reference editing with the video model, plus Qwen detail refinement.
ComfyUI-MiniMax-H3-StudioReference-aware still-image workflow built on the video model; alpha.
TE-Speed-MiniMaxH3-OSSStep cache that reuses late-block residuals when sigma barely changes, about 45% faster at 30 steps; needs a one-time ComfyUI patch.
ComfyUI-MiniMax-H3-LongMediaLong-form video and audio nodes with a Director timeline, single-clip rerenders and lip-sync redubbing.
ComfyUI-H3-ContinuumGenerates long video and audio in chunks with a prompt per chunk, lets you retry one chunk and resumes from saved takes.
ComfyUI-H3-Motion-Context-MultiRefFork of H3-Motion-Context that adds music videos built around one song, V2V motion transfer, AV bridges between two clips and masked keyframes.
comfyui-obvpm-timelineTimeline node that generates, extends, prepends, bridges and loops clips like a small video editor; needs comfyui-obvpm.
ComfyUI-YCNodes-MiniMax-H3Six nodes, including time-split prompts, dynamic CFG, low-noise refinement and tiled sampling.
Goohai-MiniMax-H3_IntegrationAll-in-one conditioning node that picks T2VA, FL2VA, Ref2VA or hybrid mode from the uploaded media and writes the official tags.
Comfyui-MMH3-UltimateUpscaleUpscales and re-samples a finished latent in time chunks and tiles, so long high-resolution clips fit limited VRAM.
ComfyUI-MAINodesMotion Lab renders a slowed copy and drops held frames to fix smeared fast motion; also turnaround and low-VRAM nodes.
ComfyUI-MiniMaxH3-PrefixStreamLong-video continuation that writes the previous clip's clean latents into the start of the next, with a clip bin for picking which take to continue.
ComfyUI-ClipProjSwaps the 32B text encoder for Qwen3-VL 4B or 8B through a learned projection, down to 4.5 GB.
ComfyUI_MinimaxH3HybridLoaderLoader that merges chosen Ref2VA tensor groups into FL2VA at load time by block range, without writing a new file.
ComfyUI-Fizgig-H3-StillTwo nodes for true single-frame stills, with a one-frame latent and a clean decode at any size; no extra model.
ComfyUI-H3-AudioRefineFreezes the video of a sampled latent and runs extra steps on the audio only, for 4-step Turbo runs whose sound is too rough.
ComfyUI-H3-Continuity-StudioMulti-shot timeline that carries context between clips and across scene or camera changes, with an MCP adapter for coding agents.
ComfyUI-sol-attnSol-Attn Triton kernel with H3 nodes for zero-copy attention, fused AdaLN modulation and chunked MLP, on RTX 30 series through Blackwell.
MiniMaxH3-StudioComfyUI sidebar with 28 nodes for prompts, the official H3 workflows, long-video planning and a built-in editor.
Veda-on-ComfyUIOne node that adds Veda sparse attention to native H3, 2.9x faster end to end on an RTX 5070, on NVIDIA, AMD ROCm and Apple silicon.
WAS Node SuiteNode suite whose H3 nodes condition a multi-segment video on one node, extend it by window and steer segments with Fun ControlNet.
ComfyUI-MiniMax-ContextIRNodes from prompt to sampling for local H3, with optional prompt rewriting through MiniMax's H3 Context-IR API.

Workflows

MiniMax-H3-Multishot-WorkflowStrings FL2VA and Ref2VA clips into one sequence with matched audio handoffs.
comfy-MiniMax-H3-workflowsThe low-VRAM acceleration stack in one importable workflow.
MiniMax-H3-8GB-WorkflowsThree-tier workflow pack for 8 GB laptops.
FL2V GGUF workflowRuns the GGUF-quantized FL2VA model in ComfyUI.
OrbitQuant workflowsT2VA workflow for the W4A4 build, with T2VA and Ref2VA API forms.
ComfyUI-MiniMax-H3-WorkflowFirst/last-frame, image-to-video and text-to-video in one workflow, with the Turbo LoRA and a control panel for resolution.

Prompting and agent skills

ComfyUI-MiniMax-H3-PromptorBuilds H3 prompts with <Picture X> tags placed inline in the action.
ComfyUI-MiniMax-H3-GuideSplits identity, keyframes, motion, voice and score into roles and compiles valid H3 prose.
OpenH3-IRWrites prompts in the Context-IR format MiniMax published, then checks and fixes them; CLI, HTTP or ComfyUI.
comfyui-minimax-h3-prompt-enhancer-T8Server-side prompt enhancement in ComfyUI.
MiniMax-H3-Prompt-Rewriter-LoRAQwen3.6-27B LoRA that rewrites short prompts into H3's three-part structure; an 8B vision version also reads keyframes.
awesome-minimax-h3-promptsPrompt collection with example videos, by category.
MiniMax H3 1K prompts1,000 curated text-to-video prompts with attribution.
minimax-h3-prompt-skill-T8Case library installable as an agent skill, with a desktop viewer.
awesome-minimax-h3 (skills)Agent skills that take a brief to a finished file through the hosted API.
okaypic-video-skillAgent skill that produces a captioned multi-shot episode through the okaypic API.
ComfyUI-Agent-KitDrives a local ComfyUI from coding agents, with an H3 skill.
minimax-h3-opencode-skillsOpenCode skills for directing, multi-shot planning, prompting and review.
MiniMax-H3-Prompt-Rewriter-ComfyUIComfyUI nodes for LightX2V's Prompt Rewriter LoRA, plus reference captioning.
ComfyUI-H3-VisionPromptorTurns an idea and optional reference images into an H3 prompt with a local vision model.
h3liteCodex skill that picks a ComfyUI setup for your hardware and generates H3 video locally.
Minimax-H3-Prompt-AgentSkillBuilds a formatted H3 prompt from your media and idea, then hands it back for you to run.
ComfyUI-Fantastic-MiniMaxH3-PromptBuilderFillable prompt templates for every mode, checked against the official guide as you type.
ComfyUI-MiniMaxH3-Prompt-WriterPrompt-writing panel in ComfyUI that turns a short brief and reference media into the official H3 format, using Ollama, a local GGUF or an API model.
minimax-h3-video-reverse-skillAgent skill that reads a reference video and writes an H3 text-to-video prompt or per-shot image-to-video prompts.
h3-storyboard-skillClaude Code skill that breaks a script into H3 shots and directs facial performance, marking each rule as tested or inferred.
oh-my-minimaxh3-directorAgent skill that splits a script into shots, batch-runs them through local ComfyUI workflows and assembles a JianYing draft.
MiniMax-H3-CloudCodex plugin that rents a GPU on the CompShare cloud, sets up ComfyUI, runs text, keyframe or reference-to-video jobs and shuts the instance down.
ComfyTVCanvas app for ComfyUI whose built-in agents ship with an H3 cinematic-director skill.
drama-skillsShort-drama skills from novel to storyboard to prompts, with an H3 mode that binds reference images and speakers.
shuohao-skillsShort-drama skills for character, art and screenplay bibles and storyboards, with the storyboard step aligning cuts and keyframes to H3 prompts.
h3-prompt-journalCase studies of music-driven dance and multi-shot prompts, each with the full prompt and what made it work.
awesome-MiniMax-H3-casesOver 3,000 playable H3 videos with their sources, over 700 complete prompts and guides for running H3 locally.
H3 Prompt ComposerOffline prompt builder in a single HTML file that runs in the browser.
Scenario skillsToolkit for agents on Scenario's API whose video skill covers H3 keyframe and reference modes and native audio prompts.
ChatCut Agent PluginAI video editor for coding agents whose video skill submits H3 and H3 Max jobs to its hosted service.
oh-story-dshDeepSeek Harness plugin for web novels and short dramas that renders video with H3 through the MiniMax API.
God MiniMax H3Prompt library and agent skill for 30-second H3 films built from chained segments, with API parameters for each prompt.
ancient-rap-battleClaude Code skill for rap battles between historical rivals, with an optional H3 lip-sync route on the MiniMax API.
short-drama-productionCodex skill that runs a short drama from outline to rough cut and submits H3 jobs to the MiniMax API or CompShare.
h3-game-spritesAgent skill that turns OpenRouter H3 clips into 2D sprite sheets, one clip per move, keyed and checked frame by frame.
MiniMax H3 B-roll GeneratorCodex skill that finds where a talk needs cutaways, writes H3 prompts and renders approved shots via MiniMax or Metaso.
Goose SkillsGrowth and marketing skills for coding agents, with a creator part that makes talking-creator clips with H3 Max on fal.
Motion Graphics Music Video SkillClaude plugin that turns a music track and a prompt into an animated video, with H3 Max image-to-video on fal.

Training

MiniMax-H3-FineTuningSupervised LoRA training on the diffusers implementation, with audio in the loss.
verl-omniRL post-training of H3 LoRAs with FlowGRPO and DiffusionNFT.
miles_diffusionFlow-GRPO post-training with an H3 text-to-audio-video recipe.
FizgigLoRA and LoKr training studio with an H3 preset.
musubi-tunerLoRA training scripts with experimental H3 support.
diffusion-pipePipeline-parallel trainer with H3 text-to-image and text-to-audio-video training.
SimpleTunerFine-tuning kit for image, video and audio models, including H3.
VeOmniByteDance's training framework with FL2VA and Ref2VA configs for H3.
UniRLTencent Hunyuan's RL framework; H3 text-to-audio-video is supported on the training side only for now.
Flow-FactoryFlow-matching fine-tuning framework with GRPO LoRA recipes for H3 text-, first/last-frame- and reference-to-audio-video.

Serving and hosting

SGLangNative H3 serving with a cookbook of recipes verified from one RTX 4090 to B300 and AMD MI355X.
vLLM-OmniOpenAI-compatible /v1/videos serving with an official H3 recipe.
UniServeHTTP serving for FastH3 and FastH3-Omni on 1 to 8 GPUs.
minimax-h3-gb300-ebpfReproducible GB300 benchmarking and tracing of SGLang H3 serving.
H3 Max on falHosted H3 Max, Director and Turbo models, post-trained by fal; closed weights, also reachable from ComfyUI Partner Nodes.
Nuva LabDedicated H3 and FastH3 hosts for one company, including customer fine-tuned weights.
MiniMax H3 on falH3 itself on fal's API, with text, image and reference to video at up to 2K.
MiniMax H3 on ReplicateH3 on Replicate's prediction API, with text, first and last frame and reference to video at 768P or 2K.
MiniMax H3 on WaveSpeedThe open H3 weights served as an API, with text, image and reference to video.
H3 on OpenRouterH3 and H3 Max video through OpenRouter's API.
MiniMax H3 on CloudflareH3 in Cloudflare's AI model catalog, with text, first and last frame and reference to video.

Guides

Run H3 locallyThis repository's guide: VRAM recipes, every quantized file, components, compatibility and measured environments.
GPU and VRAM requirementsNuva Lab's notes on GPUs and VRAM for local H3, with a cost comparison against the API.
minimax-h3-notesSingle-machine run notes.
minimax-h3-turingField handbook for 22 GB Turing cards with W4A8 and Turbo.
-h3-8gb-trapsMeasured 8 GB laptop path, with a reproduction for each trap.

MiniMax-H3-Image

MiniMax-H3-Image is the open-weight image model of the H3 family, with text-to-image and multi-image editing in one checkpoint. It generates images of about 16 megapixels directly in 6 steps, at aspect ratios from 1:3 to 3:1, and takes up to 16 reference images.

Official

MiniMax-H3-ImageOfficial repository: usage guide, showcase, local deployment and a prompt writing skill.
MiniMaxAI/MiniMax-H3-ImageOriginal weights for 16-megapixel image generation and editing with up to 16 reference images; also on ModelScope.
MiniMaxAI/MiniMax-H3-Image-PEOptional prompt expander that turns short prompts into the detailed English the model expects.
Comfy-Org/MiniMax-H3-ImageComfyUI repack for native support.
ComfyUI tutorialOfficial ComfyUI guide with text-to-image and editing templates; see also the launch post.
MiniMax APIOfficial API for text-to-image, image editing and prompt expansion; for mainland China, use platform.minimax.cn.
Hailuo AIUse H3-Image in the browser or the desktop app; for mainland China, use hailuoai.com and agent.minimax.cn.
ComfyUI API nodeBuilt-in node that generates and edits images with H3-Image through your Comfy account, with an edit template.
MiniMax CLIOfficial command line that generates images with H3-Image and edits them from up to 16 references.
MiniMax MCPOfficial server that lets agents generate images with H3-Image by default and edit from up to 16 references.

Quantizations and conversions

unsloth/MiniMax-H3-Image-FP8INT8 ConvRot and FP8 DiT, 20.3 GB each, picked automatically by Unsloth Desktop.
LSDJesus/MiniMax-H3-Image-fp8-nvfp4FP8 scaled (21 GB) and NVFP4 (12.5 GB) for ComfyUI; GGUF from Q6_K to Q2_K.
Abiray/MiniMax-H3-Image-GGUFPruned-DiT GGUF from 6.88 GB with sensitive layers kept in FP32, for stock ComfyUI-GGUF.
Hearthscale/MiniMax-H3-Image-GGUFQ8_0, Q5_K_M, Q4_K_M and Q3_K_M for stable-diffusion.cpp.
AtomicChat/MiniMax-H3-Image-GGUFGGUF tuned with a layer sensitivity scan, 7.6 to 21.4 GB, with published measurements.
lingjiu09/minimax-h3-image-ggufQ3_K_M and Q4_0, tested on an 8 GB laptop.
tsolful/Minimax_H3_Image_INT6_INT4GPTQGPTQ W6A8 (15.6 GiB) and W4A8 (12.8 GiB) for ComfyUI.
speedyrulz/minimax-h3-image-int8-w4a8-hybrid14.8 GiB INT8 and W4A8 mix for 16 GB GPUs.
PotatoForge/Minimax-H3-Image-W4A4Experimental W4A4 at 11.35 GB for older GPUs.
WaveCut/MiniMax-H3-Image-OrbitQuant-W4A8W4A8 DiT and text encoder, 27.5 GB in total.
WaveCut/MiniMax-H3-Image-PrunedUnused weights removed, 135 GB down to 95 GB with bit-identical text-to-image output; build scripts.
SD-Next SDNQ uint4SDNQ 4-bit for diffusers and SD.Next; also 4-bit and 8-bit builds by OzzyGT.
nativ-community/MiniMax-H3-Image-4bitMLX 4-bit (54.5 GB) for mlx-vlm; an 8-bit build is also available.
VisualInference/MiniMax-H3-Image-MLX-bf16MLX BF16 for the minimax-h3-image-mlx runner, with 4-bit and 8-bit builds by cguldogan.
Sawfwair/MiniMax-H3-Image-MLX-8bitMLX 8-bit export (52 GB) for mere.run.

Distillation and adapters

minimax-h3-image-pdmd-2stepTwo-step pDMD distillation LoRA, with a demo Space.
Fizgig H3-Image Training AdapterRank-16 adapter kept on while training LoRAs.
Felldude/MiniMaxImage_hdr_diffusers_vaeVAE with a fine-tuned decoder that trades pixel accuracy for more high-frequency detail, with metrics against the base VAE.

Runtimes

ComfyUINative support with official templates.
SGLangServing from the main branch, with a cookbook of recipes from DGX Spark to B300 and offload for 12 to 48 GB cards.
vLLM-OmniServing with an official recipe, including AMD Instinct.
diffusersModular Diffusers blocks for text-to-image and reference editing, released in v0.43.0; see the docs.
LightX2VText-to-image and editing with one or more references, on one GPU or several with TP/SP parallelism.
DiffSynth-StudioInference, low-VRAM inference, and full or LoRA training.
stable-diffusion.cppC/C++ inference with day-0 H3-Image support and GGUF models.
UnslothRuns H3-Image in its desktop app and web UI with the INT8 or FP8 DiT.
mmh3Rust engine with CUDA and Metal kernels.
h3-image-mlxMLX generation and editing on M1 to M4 Macs with W8A8 weights, about 7 GB peak memory at 4K.
minimax-h3-image-mlxBF16 MLX port that reads each block from disk while the previous one computes; 5 GB of memory at 2048×2048.
TensorSharp.NET engine that runs H3-Image text-to-image on ggml; editing is not implemented yet.
minimax-h3-image-serverServer with a browser UI for generation and reference edits; about 21 GB of VRAM by default, down to 8 GB with W4A8 and weight streaming.
MiniMax-H3-Image-AccelSGLang tuned for L20 and RTX 5090, BF16 by default with optional FP8 for lower latency.
mlx-vlmMLX text-to-image and image editing for H3-Image on Apple silicon, with BF16, 8-bit and 4-bit checkpoints.

Apps

SwarmUIWeb UI that runs H3-Image with up to 16 prompt images for references or edits.
PallaidiumBlender video editor add-on with H3-Image text-to-image and edits from up to 10 references, in SDNQ 4-bit for 24 GB GPUs.
diffusion-for-kritaKrita plugin on a ComfyUI backend that uses H3-Image to generate, inpaint and edit with up to 16 images.
styk-paintGodot add-on that has H3-Image repaint a 3D blockout view by view through ComfyUI, then projects the paint back onto the set.
mmh3-studiomacOS and browser client for a resident mmh3 server, so generations skip the cold start.
h3-image-editorWindows workspace on ComfyUI for H3-Image generation, reference edits and masked repair, with Qwen, SAM and SeedVR2 tools.
ImageGenerationServerMulti-user web front end for a local ComfyUI, with H3-Image generation and edits from up to 16 pictures; non-commercial license.
tongflow-modal-minimax-h3-imageTongFlow plugin that runs H3-Image generation, editing and fusion of up to 16 references on Modal.

API integrations

HyperFramesHTML-to-video renderer for agents whose media command can make 2K or 4K images with H3-Image through HeyGen.
AI SDKVercel's TypeScript toolkit whose MiniMax provider generates and edits images with H3-Image and its Highspeed version.
RevericDesktop studio for novels and short dramas that creates and edits images with H3-Image through the MiniMax API.
Dify MiniMax Image pluginTool for Dify apps that generates images with H3-Image and edits from 1 to 16 references through the MiniMax API.
OpenDirectorAgent video workflow tool with an H3-Image adapter that edits from up to 16 references through the MiniMax API.
Octafuse GatewaySelf-hosted API hub with presets for H3-Image generation and edits on MiniMax's OpenAI-compatible Images API.

ComfyUI nodes and workflows

ComfyUI-MiniMaxH3-ToolsNodes that free the text encoder before the DiT loads, so H3-Image runs on a 24 GB GPU with 32 GB of RAM.
ComfyUI-MiniMaxH3-EasyIts Easy Image node does H3-Image text-to-image and reference edits; type @ in the prompt to insert a reference.
ComfyUI-H3Image-MediaLoaderLoads up to 16 reference images with per-slot crop, rotate and reorder.
H3-Video-LoRA-Loader-for-ImageRuns H3 video LoRAs on H3-Image.
ComfyUI-H3-Continuity-StudioNode pack for multi-clip H3 video, with an H3-Image node that makes the still frame guiding each shot.
comfyui-minimax-h3-prompt-enhancer-T8Prompt enhancement nodes, with H3-Image support since v1.41.0.
ComfyUI-OrbitQuantRuns the OrbitQuant W4A8 build for generation and edits with up to 16 references.
H3GIFHas H3-Image draw a storyboard grid in one image, then splits it into frames for a looping GIF.
ComfiestAgent-driven ComfyUI workflow suite with H3-Image text-to-image and edit graphs for 1, 2 or 4 references.
minimax-h3-colab-workflowColab notebooks that run H3 and H3-Image from open weights on an L4, through ComfyUI quants and stable-diffusion.cpp.
comfyui-minimax-imageRunPod template that boots into ComfyUI with the H3-Image models.
minimax-h3-v100-guideTested setup for H3 and H3-Image on one 16 GB V100 with 32 GB of RAM: configs, launchers and workflows.

Training

FizgigLoRA and fine-tune studio with H3-Image support since v9.6.
ai-toolkitGUI and CLI trainer; trains H3-Image editing when the dataset has control images.

Prompting

minimax-h3-image-promptsReusable skill for writing generation and editing prompts with up to 16 references.
awesome-minimax-h3-image-prompts63 source-linked H3-Image examples with result images and editing references; only 5 include the prompt itself.

Benchmarks

h3-image-bench600 graded prompts in two rounds at about 16 MP on a 16 GB RTX 5080, with a live results page.
M5 Pro Mac mini benchmarkComfyUI INT8 timings on a 64 GB M5 Pro Mac mini: about 9 minutes per 4096×4096 image.
12 GB testsINT8 on a 12 GB RTX 4070 SUPER with 96 GB of RAM: 11 to 12 s at 4 MP, about 41 s at 16 MP.

Contributing

Add one row to the matching table: the owner's avatar, the name linked to the project, and one sentence on what it is. Add the same row to README_cn.md, then run python3 scripts/check_readme.py.

An entry needs its own README, docs or model card that says what it is and that it uses MiniMax H3 or MiniMax-H3-Image, locally or through an API. Measurements, VRAM recipes and per-file details belong in docs/run-h3-locally.md.

Acknowledgements

This index is only possible because other people made the models, tools, tests, and documentation it points to. Thank you to the MiniMax, ComfyUI, SGLang, vLLM, NVIDIA, and Unsloth teams, and to the independent maintainers who keep testing H3 on real hardware.

The structure and much of the resource discovery come from the community-maintained wildminder/awesome-minimax-H3. This page follows that work and adds MiniMax's own GitHub and Hugging Face scan.

More specific thanks go to:

  • Comfy-Org for the official ComfyUI conversions, workflow templates, and day-one support.
  • ModelTC / LightX2V for the Turbo distillation work and public training configuration.
  • Larryvrh for the checkpoint comparisons behind the Turbo guidance.
  • Kijai for the NVIDIA Sol-Attn implementation and benchmark notes.
  • Alibaba PAI and aigc-apps/VideoX-Fun for the PDD 8-step Acc LoRAs and the Fun ControlNet-Union branch.
  • OpenVDN for publishing Video Delta Net's training code alongside the weights.
  • FastVideo for the FastH3 line (4-step preview, V2, Trim and Omni), its data-free distillation recipe, and UniServe.
  • Nuva Lab for the production grounding behind FastH3, the FastH3-Omni co-development and reference dataset, and the production hardware behind the UniServe benchmarks.
  • Pika Labs for the six-step Flex adapters and inference examples; NVIDIA for the released draft refiner and fast decoder, building on LBH, LightX2V and LMS components.
  • IAmIronMan42 for the training work and the documented fixes.
  • Salvatore Sanfilippo (antirez) for the standalone H3 C/Metal inference engine.
  • Every quantizer and workflow maintainer represented above. Their files, testing time, and write-ups make local H3 use much easier.

If you spot a wrong number, broken link, or missing compatibility note, please open an issue or send a correction.

Contact

For MiniMax H3 questions, contact model@minimax.io.

关于 About

No description, website, or topics provided.

语言 Languages

Python100.0%

提交活跃度 Commit Activity

代码提交热力图
过去 52 周的开发活跃度
74
Total Commits
峰值: 20次/周
Less
More

核心贡献者 Contributors