---
license: apache-2.0
language:
  - en
library_name: transformers
pipeline_tag: text-generation
tags:
  - moe
  - mixture-of-experts
  - gravity
  - txgravity
  - therapeutics
  - tdc
  - trillion-labs
base_model:
  - trillionlabs/Gravity-30B-A5B
---

# TxGravity-30B-A6B-Open

## Model Summary

**TxGravity-30B-A6B-Open** is a therapeutics-focused, task-specialized model built on the **Gravity-30B-A5B** base. It predicts a broad range of therapeutic properties — small-molecule ADMET, toxicity, drug–target interaction, protein–protein and peptide–MHC interaction, and more — following the instruction format of the **Therapeutic Data Commons (TDC)** benchmark, in the same spirit as Google's TxGemma.

It is trained on TDC therapeutic instruction data (single-turn instruction → answer) covering 57 tasks. Answers are formatted as `(A)`/`(B)` for binary classification or a normalized `000`–`1000` bin for regression.

| Property | Value |
|---|---|
| Total Parameters | 29.56B |
| Active Parameters | ~6B |
| Architecture | GravityMoE (DeepSeek-V3-compatible) |
| Layers | 52 |
| Routed / Shared Experts | 64 (top-8) / 1 |
| Context Length | 8,192 tokens |
| Precision | bf16 |
| Base model | Gravity-30B-A5B |

> ⚠️ This is a task-specialized property predictor, not a general instruction-tuned or safety-aligned assistant. Its outputs are intended for the TDC-style therapeutic prediction prompts it was trained on. Predictions may be inaccurate, biased, or incomplete and must be independently verified before any experimental, clinical, or decision-making use.

## Performance

Evaluated on the TDC therapeutic benchmark against a reproduced **TxGemma-27B** baseline across 60 comparable tasks:

**16 wins · 26 ties · 18 losses**

Selected results (TxGravity vs TxGemma-27B):

| Task | Metric | TxGravity | TxGemma-27B |
|---|---|---|---|
| bindingdb_ki | PCC ↑ | **0.750** | −0.112 |
| bindingdb_ic50 | Spearman ↑ | **0.766** | 0.643 |
| protein_sabdab | MAE ↓ | **0.969** | 2.332 |
| ld50_zhu | MAE ↓ | **0.651** | 0.776 |
| drugcomb_css | MAE ↓ | **7.838** | 9.418 |
| bioavailability_ma | AUROC ↑ | **0.749** | 0.697 |
| bbb_martins | AUROC ↑ | **0.919** | 0.897 |
| ames | AUROC ↑ | **0.841** | 0.830 |
| davis | MSE ↓ | 0.760 | **0.643** |
| kiba | MSE ↓ | 0.637 | **0.460** |

Full per-task results (66 tasks, metric, N, margin, verdict) are in [`summary.csv`](./summary.csv).

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
tok = AutoTokenizer.from_pretrained("trillionlabs/TxGravity-30B-A6B-Open")
model = AutoModelForCausalLM.from_pretrained(
    "trillionlabs/TxGravity-30B-A6B-Open", dtype="bfloat16", device_map="auto")

messages = [{"role": "user", "content": "<TDC-formatted instruction here>"}]
ids = tok.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
out = model.generate(ids, max_new_tokens=16)
print(tok.decode(out[0][ids.shape[1]:], skip_special_tokens=True))
```

The model expects TDC-style prompts and answers in TDC format: `(A)`/`(B)` for classification, or a `000`–`1000` normalized bin for regression.

## License
Apache 2.0