clausify-models is an open-weight model from AFNAN MAZUMDER, released under Apache License 2.0. Its published files total 1.1 GB.
Two fine-tuned distilbert-base-uncased checkpoints used by the Clausify app (https://github.com/afnanmz168/clausify-p3-app), trained on the 408-contract training split of CUAD (seed-42, contract-level 80/20 split). Research prototype — not legal advice.
Model Card
By AFNAN MAZUMDER, published under apache-2.0, revision 763d71b59fda.
Two fine-tuned distilbert-base-uncased checkpoints used by the Clausify app (https://github.com/afnanmz168/clausify-p3-app), trained on the 408-contract training split of CUAD (seed-42, contract-level 80/20 split). Research prototype — not legal advice.
Read AFNAN MAZUMDER's full model card
Clausify models (CUAD)
Two fine-tuned distilbert-base-uncased checkpoints used by the Clausify app
(https://github.com/afnanmz168/clausify-p3-app), trained on the 408-contract training split of
CUAD (seed-42, contract-level 80/20 split). Research prototype — not legal advice.
| Folder | Task | Held-out result (102 test contracts) |
|---|---|---|
presence_mil/final |
(category question, 2,000-char window) → does the window contain the clause? Max-pooled over windows. | micro-F1 0.692 alone; 0.779 in an AND-ensemble with a TF-IDF baseline |
span/final |
Extractive QA: clause span for a category question | token-F1 0.764 given an answer-bearing window |
Details and full evaluation: https://github.com/afnanmz168/clausify-p3
Identity and Version
- Repository
- af123Af/clausify-models
- Publisher
- AFNAN MAZUMDER
- Task
- Not stated by the source
- Modality
- Other
- Library
- transformers
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 763d71b59fda2591eabb122fcfde06ba3533d921
- First published
- 2026-10-04
- Last updated
- 2026-10-09
Files and Weights
21 files, 1.1 GB in total. The weights are 4 files totalling 1.1 GB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| presence_mil/final/model.safetensors | Weights | 267.8 MB | a5482a0302ae |
| presence_v2/final/model.safetensors | Weights | 267.8 MB | e5f84bd42eb8 |
| span/final/model.safetensors | Weights | 265.5 MB | 380a233a3c03 |
| summarizer/final/model.safetensors | Weights | 307.9 MB | 9fcb6bdf777d |
| presence_mil/final/config.json | Configuration | 664 B | — |
| presence_v2/final/config.json | Configuration | 664 B | — |
| span/final/config.json | Configuration | 610 B | — |
| summarizer/final/config.json | Configuration | 1.6 KB | — |
| summarizer/final/generation_config.json | Configuration | 152 B | — |
| README.md | Documentation | 909 B | — |
| baseline/baseline.pkl | Other | 7.1 MB | 903e52750040 |
| baseline_v2/baseline.pkl | Other | 7.1 MB | 3fc0d273065d |
| .gitattributes | Repository | 1.5 KB | — |
| presence_mil/final/tokenizer.json | Tokenizer | 711.5 KB | — |
| presence_mil/final/tokenizer_config.json | Tokenizer | 351 B | — |
| presence_v2/final/tokenizer.json | Tokenizer | 711.5 KB | — |
| presence_v2/final/tokenizer_config.json | Tokenizer | 351 B | — |
| span/final/tokenizer.json | Tokenizer | 711.5 KB | — |
| span/final/tokenizer_config.json | Tokenizer | 351 B | — |
| summarizer/final/tokenizer.json | Tokenizer | 2.4 MB | — |
| summarizer/final/tokenizer_config.json | Tokenizer | 2.4 KB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 1.1 GB
Released by AFNAN MAZUMDER through its official repository on Hugging Face. Read the license.
Built From
- Derived from distilbert/distilbert-base-uncased
- Trained on (disclosed) theatticusproject/cuad
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 1.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About clausify-models
Can I use clausify-models commercially?
Yes. clausify-models is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.