This repository hosts pretrained EdenGNN checkpoints and reference configuration files associated with the paper Efficient equivariant framework for universal charge density prediction. EdenGNN (Equivariant Density Graph Neural Network) predicts charge densities from atomic structures and can be coupled with downstream DFT workflows for electronic-structure calculations. The config.yaml files stored in this repository are for reference only. They are not plug-and-play configuration files and must be modified based on the user's own environment before use. In particular, users should update paths, software-specific templates, dataset locations, checkpoint paths, output directories, and…
A trainable knowledge graph embedding model encoding the unified evidence hypergraph for Case 2025-137857.
Parameters35M
Context—
Weights144.2 MB
Licensemit
AccessOpen weights
Monthly Downloads162
Runs On
What it takes to serve unicosys-hypergraph (35M 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 | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.0 GB | 0.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.0 GB | 0.0 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 Sep 18, 2026.
Model Card
By Dan Faucitt, published under mit, revision 386bbd309d7f.
A trainable knowledge graph embedding model encoding the unified evidence hypergraph for Case 2025-137857. This model encodes a unified hypergraph linking financial transactions, email communications, legal evidence, and entity relationships into a single trainable knowledge representation. The model can be fine-tuned on link prediction tasks: - model.safetensors — Model weights - config.json — Model configuration - graphdata.safetensors — Encoded graph tensors (nodes, edges) - tokenizer.json — Character-level tokenizer for node labels - nodeidmapping.json — Node ID string to integer index mapping - modelsummary.json — Compact statistics summary Generated by the Unicosys intelligence…
Read Dan Faucitt's full model card
Unicosys Hypergraph Knowledge Model
A trainable knowledge graph embedding model encoding the unified evidence hypergraph for Case 2025-137857.
Model Description
This model encodes a unified hypergraph linking financial transactions, email communications, legal evidence, and entity relationships into a single trainable knowledge representation.
Architecture
| Component | Details |
|---|---|
| Node Embedding | 128-dim structural + 256-dim text |
| Hidden Dimension | 256 |
| Text Encoder | 2-layer Transformer, 4 heads |
| Graph Attention | 2-layer GAT, 4 heads |
| Link Predictor | 2-layer MLP with margin ranking loss |
| Total Parameters | 34,762,497 |
Knowledge Graph Statistics
| Metric | Count |
|---|---|
| Total Nodes | 198,019 |
| Total Edges | 13,415 |
| Cross-Links | 3,664 |
| Entities | 16 |
| Emails | 197,993 |
| Financial Documents | 0 |
| Timeline Events | 10 |
| LEX Schemes | 0 |
| Legal Filings | 0 |
Subsystems
| Subsystem | Nodes |
|---|---|
| Core (Entities) | 16 |
| Fincosys (Financial) | 0 |
| Comcosys (Communications) | 197,993 |
| RevStream1 (Evidence) | 0 |
| Ad-Res-J7 (Legal) | 10 |
Training
The model can be fine-tuned on link prediction tasks:
from model.unicosys_model import UnicosysHypergraphModel, UnicosysConfig
model = UnicosysHypergraphModel.from_pretrained("hyperholmes/unicosys-hypergraph")
# ... prepare training data ...
# model.forward(node_ids, node_type_ids, subsystem_ids, edge_index, edge_type_ids,
# pos_edge_index=pos, neg_edge_index=neg, labels=labels)
Files
model.safetensors— Model weightsconfig.json— Model configurationgraph_data.safetensors— Encoded graph tensors (nodes, edges)tokenizer.json— Character-level tokenizer for node labelsnode_id_mapping.json— Node ID string to integer index mappingmodel_summary.json— Compact statistics summary
Source
Generated by the Unicosys intelligence pipeline.
Configuration
- Architecture
- UnicosysHypergraphModel
- Model type
- unicosys_hypergraph
Identity and Version
- Repository
- drzo/unicosys-hypergraph
- Publisher
- Dan Faucitt
- Task
- Graph ml
- Modality
- Other
- Library
- transformers
- Parameters
- 35M parameters
- Languages
- en
- Revision
- 386bbd309d7f17dfe98efb3973c7c01f245d74e5
- First published
- 2026-03-14
- Last updated
- 2026-09-18
Files and Weights
11 files, 152.2 MB in total. The weights are 2 files totalling 144.2 MB in safetensors.
Weights2 files · 144.2 MB
Configuration6 files · 8.0 MB
Tokenizer1 file · 2.5 KB
Documentation1 file · 2.2 KB
Repository1 file · 1.5 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| graph_data.safetensors | Weights | 5.1 MB | bb53e0250a42 |
| model.safetensors | Weights | 139.1 MB | 12185733ff81 |
| __init__.py | Configuration | 177 B | — |
| config.json | Configuration | 980 B | — |
| configuration_unicosys.py | Configuration | 2.3 KB | — |
| model_summary.json | Configuration | 566 B | — |
| modeling_unicosys.py | Configuration | 11.8 KB | — |
| node_id_mapping.json | Configuration | 8.0 MB | — |
| README.md | Documentation | 2.2 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
| tokenizer.json | Tokenizer | 2.5 KB | — |
License and Download
- License
- mit
- Access
- Open weights, no gate
- Download size
- 144.2 MB
Download from Dan Faucitt
Released by Dan Faucitt through its official repository on Hugging Face. Read the license.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 144.2 MB |
| 16-bit | 0.1 GB |
| 8-bit | 0.0 GB |
| 4-bit | 0.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About unicosys-hypergraph
How much GPU memory does unicosys-hypergraph need?
About 0.1 GB at 16-bit and 0 GB at 4-bit: the weights (35M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run unicosys-hypergraph 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 unicosys-hypergraph commercially?
Yes. unicosys-hypergraph is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.