Open-weight model
unimate-uniml3d-f60-v3-cross-attn-bf16
by MLX Community mlx-community/unimate-uniml3d-f60-v3-cross-attn-bf16
unimate-uniml3d-f60-v3-cross-attn-bf16 is an open-weight model from MLX Community, released under Creative Commons Attribution-NonCommercial 4.0. It has 176M parameters. At 16-bit it needs about 0.4 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
UniMate (SIGGRAPH Asia 2026) generates motion from text for skeletons of any topology: animals, humanoids and rigged objects.
Runs On
What it takes to serve unimate-uniml3d-f60-v3-cross-attn-bf16 (176M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.4 GB | 0.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.2 GB | 0.2 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 18, 2026.
Model Card
UniMate (SIGGRAPH Asia 2026) generates motion from text for skeletons of any topology: animals, humanoids and rigged objects. A flow-matching transformer, conditioned on a flan-t5-base encoding of the prompt and on the skeleton's T-pose, hierarchy and joint names, turns noise into 60 frames (2 s at 30 fps) of motion. This repository holds the unimateuniml3df60v3crossattn checkpoint converted to MLX for Apple silicon, for use with mlx-vlm. mlx-community/unimate-uniml3d-f60-v3-bf16 holds the recommended graph attention x adaLN model (unimateuniml3df60v3), converted the same way. This is an unofficial conversion, not made or endorsed by the UniMate - Extracted the EMA weights…
Excerpt from the card by MLX Community, licensed cc-by-nc-4.0.
Configuration
- Vocabulary size
- 32,128
- Model type
- unimate
Identity and Version
- Repository
- mlx-community/unimate-uniml3d-f60-v3-cross-attn-bf16
- Publisher
- MLX Community
- Task
- Not stated by the source
- Modality
- Other
- Library
- mlx
- Parameters
- 176M parameters
- Languages
- en
- Revision
- fb847227f259d2db07e12f9a2b2f0397f9a77686
- First published
- 2026-10-14
- Last updated
- 2026-10-14
Files and Weights
10 files, 354.9 MB in total. The weights are 1 file totalling 351.6 MB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 351.6 MB | 07cc65fec628 |
| config.json | Configuration | 2.0 KB | — |
| model.safetensors.index.json | Configuration | 31.3 KB | — |
| special_tokens_map.json | Configuration | 2.2 KB | — |
| LICENSE | Documentation | 20.0 KB | — |
| README.md | Documentation | 10.1 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
| spiece.model | Tokenizer | 791.7 KB | d60acb128cf7 |
| tokenizer.json | Tokenizer | 2.4 MB | — |
| tokenizer_config.json | Tokenizer | 2.5 KB | — |
License and Download
- License
- cc-by-nc-4.0
- Access
- Open weights, no gate
- Download size
- 351.6 MB
Released by MLX Community through its official repository on Hugging Face. Read the license.
Built From
- Derived from Linzhan/UniMate
- Described by arXiv:2609.05415
- Trained on (disclosed) Linzhan/UniML3D
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 351.6 MB |
| 16-bit | 0.4 GB |
| 8-bit | 0.2 GB |
| 4-bit | 0.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About unimate-uniml3d-f60-v3-cross-attn-bf16
How much GPU memory does unimate-uniml3d-f60-v3-cross-attn-bf16 need?
About 0.4 GB at 16-bit and 0.1 GB at 4-bit: the weights (176M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run unimate-uniml3d-f60-v3-cross-attn-bf16 on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use unimate-uniml3d-f60-v3-cross-attn-bf16 commercially?
Not without separate permission. unimate-uniml3d-f60-v3-cross-attn-bf16 is released under Creative Commons Attribution-NonCommercial 4.0. CC BY-NC 4.0 permits sharing and adapting with credit for non-commercial purposes only. Commercial use needs separate permission from the rights holder.