Unverified50% confidenceFactExact time
The model used in this workflow is described as minimax ref2va pruned int8 convrot with turbo 4 step lora.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
UnverifiedMiniMax adopts a Transformer architecture combined with a Diffusion Model.63% similarUnverifiedMiniMax H3 is a multimodal generative model with two core inference modes: REF2VA and FL2VA.61% similarUnverifiedThe LightX2V team released a new version of Minimax-h3-Turbo on Hugging Face — FL2V Turbo 4-step v1.2 (768p) — an accelerated LoRA that claims to complete video generation in just 4 inference steps.59% similarUnverifiedMiniMax is building a Pro frontier flagship model at the 2.7-trillion-parameter scale.58% similarUnverifiedThis test used the minimax_h3_fl2va_int8_convrot model running on a ComfyUI workflow.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/909162API
curl https://kongchang.com/api/v1/knowledge/claims/909162MCP
get_claim(id=909162)