初始化项目，由ModelHub XC社区提供模型

Model: QizhiPei/BioMatrix-1.7B-SFT
Source: Original Platform

.gitattributes

@@ -0,0 +1,42 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+figures/3D_tokenizers.png filter=lfs diff=lfs merge=lfs -text
+figures/biomatrix_arch_v1.png filter=lfs diff=lfs merge=lfs -text
+figures/biomatrix_cpt_data_v5.png filter=lfs diff=lfs merge=lfs -text
+figures/biomatrix_logo.png filter=lfs diff=lfs merge=lfs -text
+figures/molecule_tasks_v1.png filter=lfs diff=lfs merge=lfs -text
+figures/protein_tasks_v1.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,213 @@
+---
+license: apache-2.0
+language:
+- en
+tags:
+- biology
+- chemistry
+- molecule
+- protein
+- multimodal
+- foundation-model
+- drug-discovery
+- protein-design
+pipeline_tag: text-generation
+base_model: QizhiPei/BioMatrix-1.7B-Base
+library_name: transformers
+---
+
+# BioMatrix-1.7B-SFT
+
+<div align="center">
+  <img src="figures/biomatrix_logo.png" width="60%" alt="BioMatrix" />
+</div>
+
+**BioMatrix** is a multimodal biological foundation model that natively integrates **1D sequences**, **3D structures**, and **natural language** for both **molecules** and **proteins** within a single decoder-only architecture.
+
+This is the **1.7B-parameter SFT (Supervised Fine-Tuned)** variant, instruction-tuned across 80 downstream biological tasks spanning 6 categories. For a larger and more capable model, see [BioMatrix-4B-SFT](https://huggingface.co/QizhiPei/BioMatrix-4B-SFT).
+
+- 📄 **Paper**: [BioMatrix: Towards a Comprehensive Biological Foundation Model Spanning the Modality Matrix of Sequences, Structures, and Language](http://arxiv.org/abs/2606.22138)
+- 💻 **Code**: [https://github.com/QizhiPei/BioMatrix](https://github.com/QizhiPei/BioMatrix)
+- 🤗 **Model & Data Collection**: [https://huggingface.co/collections/QizhiPei/biomatrix](https://huggingface.co/collections/QizhiPei/biomatrix)
+
+<p align="center">
+  <img width="90%" src="figures/biomatrix_arch_v1.png" alt="BioMatrix Architecture">
+</p>
+
+## Model Overview
+
+BioMatrix closes the gap between native multimodality and broad entity coverage in biological foundation models. Unlike adapter-based approaches that bolt external encoders onto a language model, or prior native-tokenization models confined to a single entity type, BioMatrix maps **all modalities into a shared discrete token space** via a unified tokenization scheme:
+
+- **Molecular 1D sequences** (both SMILES and SELFIES notations)
+- **Molecular 3D structures** (via MolStrucTok with branch-decoupled decoder)
+- **Protein 1D sequences** (residue-level tokens)
+- **Protein 3D structures** (via GCP-VQVAE backbone tokenizer)
+- **Natural language** (inherited from Qwen3 tokenizer)
+
+All modalities are consumed and produced uniformly under a **single next-token prediction objective**—without external encoders, projection adapters, or modality-specific output heads.
+
+| Model | Molecule 1D | Molecule 3D | Protein 1D | Protein 3D | Natural Language |
+|-------|:-----------:|:-----------:|:----------:|:----------:|:----------------:|
+| ESM3 | ✗ | ✗ | ✓ | ✓ | ✓ |
+| 3D-MoLM | ✓ | ✓ | ✗ | ✗ | ✓ |
+| AlphaFold3 | ✓ | ✓ | ✓ | ✓ | ✗ |
+| BioT5/BioT5+ | ✓ | ✗ | ✓ | ✗ | ✓ |
+| BioMedGPT | ✓ | ✗ | ✓ | ✗ | ✓ |
+| NatureLM | ✓ | ✗ | ✓ | ✗ | ✓ |
+| SciReasoner | ✓ | ✗ | ✓ | ✗ | ✓ |
+| **BioMatrix** | **✓** | **✓** | **✓** | **✓** | **✓** |
+
+<p align="center">
+  <img width="95%" src="figures/3D_tokenizers.png" alt="Molecule and Protein Structure Tokenizers">
+</p>
+
+## Model Details
+
+- **Base Architecture**: Qwen3-1.7B-Base
+- **Parameters**: 1.7B
+- **Training Stages**:
+  - **Continual Pretraining** on 304.4B tokens (general/scientific text, molecular & protein 1D/3D data, cross-modal interleaved corpora)
+  - **Instruction Tuning** on a comprehensive suite of 80 downstream tasks across 6 categories
+- **Context Length**: 8,192 tokens
+- **Tokenizer**: Extended Qwen3 vocabulary with:
+  - 11,294 joint molecular 3D tokens (composed from SELFIES atom × MolStrucTok codes)
+  - 4,096 protein 3D tokens (GCP-VQVAE codebook)
+  - 26 protein 1D tokens (amino acids + non-standard/unknown)
+  - SELFIES atom tokens and modality-specific control tokens
+
+## Pretraining Corpus (304.4B tokens)
+
+| Category | Tokens | Sources |
+|----------|--------|---------|
+| **Text** | 105.3B | FineWeb-Edu, FineFineWeb (biology/chemistry/medical/health), PubMed Full Articles |
+| **Molecule** | 73.7B | PubChem, PCQM4Mv2, PubChemQC, MolTextNet |
+| **Protein** | 77.4B | UniRef50, RCSB PDB, Swiss-Prot, TrEMBL, AlphaFold DB |
+| **Cross-entity** | 48.0B | Interleaved text (PubMed, bioRxiv, S2ORC, USPTO), Molecule–protein (BindingDB, STITCH, jglaser, CrossDocked), Protein–protein (AlphaSeq, PPIRef) |
+
+<p align="center">
+  <img width="95%" src="figures/biomatrix_cpt_data_v5.png" alt="BioMatrix Continual Pretraining Data">
+</p>
+
+## Performance Highlights
+
+Despite its compact 1.7B size, BioMatrix delivers strong performance across diverse biological tasks—often surpassing models several times larger. Selected highlights:
+
+### Molecular Tasks
+- **Unconditional 1D Generation** (GuacaMol, SELFIES): 0.999 validity, 1.000 uniqueness
+- **Name Conversion (I2S EM)**: 87.22% (surpasses SciReasoner-8B at 84.40%)
+- **Text-Based Molecule Generation (EM)**: 56.35% (vs. SciReasoner-8B: 48.00%)
+- **MoleculeQA Total**: 70.07% (vs. prior best MolCA-1.3B: 64.79%)
+- **Property-Conditioned 3D Generation**: ~3-4× error reduction on QM9 electronic-structure targets
+
+### Protein Tasks
+- **Fold Type Prediction (Family level)**: 85.84% accuracy
+- **EC Number Prediction (Price split, F1)**: 34.34% (surpasses SciReasoner-8B at 22.00%)
+- **Inverse Folding AAR**: 75.20% (vs. DPLM-2-3B: 61.67%)
+- **Sequence–Structure Co-generation**: scTM = 0.965, scRMSD = 2.81
+
+### Interaction Tasks
+- **BindingDB Affinity (RMSE)**: 1.268
+- **PDBBindv2020 3D Affinity**: best Spearman correlation (0.717) among all baselines
+
+## Quick Start
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_name = "QizhiPei/BioMatrix-1.7B-SFT"
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto",
+    device_map="auto",
+    trust_remote_code=True
+)
+
+# Example: Molecule captioning with SELFIES input
+instruction = "I need a brief explanation of the molecule denoted in this SELFIES notation. <|mol_sfi_start|>[Te]<|mol_sfi_end|>"
+
+messages = [
+    {"role": "user", "content": instruction}
+]
+prompt = tokenizer.apply_chat_template(
+    messages,
+    tokenize=False,
+    add_generation_prompt=True
+)
+
+inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+outputs = model.generate(**inputs, max_new_tokens=2048, do_sample=False)
+response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=False)
+print(response)
+```
+
+## Modality Wrapping
+
+When constructing prompts, biomolecular content must be wrapped with the corresponding control tokens:
+
+| Modality | Wrapping Example |
+|----------|------------------|
+| Molecule SMILES | `<\|mol_smi_start\|>CC#CC#N<\|mol_smi_end\|>` |
+| Molecule SELFIES | `<\|mol_sfi_start\|>[C][#C][C][#N]<\|mol_sfi_end\|>` |
+| Molecule 3D | `<\|mol_3d_start\|>[H 3][C 0][#C 6]...<\|mol_3d_end\|>` |
+| Protein 1D | `<\|prot_aa_start\|><A M><A R><A A>...<\|prot_aa_end\|>` |
+| Protein 3D | `<\|prot_3d_start\|><S 4012><S 153><S 2091>...<\|prot_3d_end\|>` |
+
+Natural language text is left unwrapped and serves as the default carrier modality.
+
+## Supported Tasks
+
+BioMatrix-1.7B-SFT was instruction-tuned across the following task categories:
+
+**Molecule (1D)**: unconditional generation, name conversion, property prediction, captioning, text-based generation, forward/retrosynthesis, editing, optimization, customized generation, question answering
+
+**Molecule (3D)**: unconditional generation, property-conditioned generation
+
+**Protein (1D)**: sequence understanding, annotation prediction, knowledge mining, text-based design, unconditional generation
+
+**Protein (3D)**: structure understanding, folding, inverse folding, sequence-structure co-generation, unconditional backbone generation
+
+**Interaction**: molecule-protein binding affinity (1D & 3D), protein-protein interaction
+
+<p align="center">
+  <img width="95%" src="figures/molecule_tasks_v1.png" alt="BioMatrix Molecule Tasks">
+</p>
+
+<p align="center">
+  <img width="95%" src="figures/protein_tasks_v1.png" alt="BioMatrix Protein Tasks">
+</p>
+
+> **Note on task-group variants**: As detailed in the paper, the released SFT model is trained on the union of all sub-task corpora with mild oversampling for small-data tasks. For best performance on specific benchmarks, please refer to the paper's task-group-specific variants.
+
+## SMILES vs. SELFIES
+
+BioMatrix supports both notations as parallel 1D molecular representations. Empirically:
+
+- **SELFIES** excels on tasks requiring validity-by-construction (unconditional generation, property optimization)
+- **SMILES** excels on tasks requiring surface-level structural anchoring (customized generation with atom/bond/functional-group constraints, forward synthesis, retrosynthesis)
+
+See Section 9.2 of the paper for detailed analysis.
+
+## Limitations
+
+- Molecular and protein 3D structures are tokenized in **disjoint geometric reference frames**, so the model cannot natively represent biomolecular complexes (e.g., docking poses).
+- Heavy domain specialization may erode some general-purpose language capabilities of the underlying Qwen3 backbone.
+- Coverage is limited to **small molecules and proteins**; nucleic acids, carbohydrates, and lipids are not currently supported.
+- Fine-grained 3D geometry (e.g., bond lengths) shows residual quantization error from finite codebooks; a lightweight post-hoc force-field refinement (e.g., MMFF) closes most of this gap.
+
+## Citation
+
+If you find BioMatrix useful, please cite:
+
+```bibtex
+@article{pei2026biomatrix,
+  title={BioMatrix: Towards a Comprehensive Biological Foundation Model Spanning the Modality Matrix of Sequences, Structures, and Language},
+  author={Pei, Qizhi and Zhou, Zhimeng and Duan, Yi and Zhao, Yiyang and Li, Wei and Guo, Han and He, Liang and Li, Chengping and Hsieh, Chang-Yu and He, Conghui and Yan, Rui and Wu, Lijun},
+  journal={arXiv preprint arXiv:2606.22138},
+  year={2026}
+}
+```
+
+## License
+
+This model is released under the Apache 2.0 license. The base model (Qwen3-1.7B-Base) is subject to its own license terms.

config.json

@@ -0,0 +1,30 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 6144,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 168064
+}

figures/3D_tokenizers.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e007796dfda9ed5badb1287bbe6a79905526a812c2090f143385c8edee7e7161
+size 849529

figures/biomatrix_arch_v1.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f47ba2590471f3bcbf28c7fd1ba978891c00de0a6ea8aed8a508d594143edae
+size 833624

figures/biomatrix_cpt_data_v5.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41ff1faa16f44286b388db0d8463dd9d51263db6f23ce99b429d53e9ea49b575
+size 511512

figures/biomatrix_logo.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2eb845f564b1f8d6b3ad0caee11d70977103f2ccb058d8aabef5c0ed7dcaf0a3
+size 1197085

figures/molecule_tasks_v1.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c800920a4829cdb5f6b920b822a7a56f48e080047232fa9df574c566ea543e4
+size 881792

figures/protein_tasks_v1.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f45f3e5d774c516f137fe935da690653a05e47837747d07467226c06b0694ed8
+size 934114

generation_config.json

@@ -0,0 +1,5 @@
+{
+  "eos_token_id": 151645,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.51.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d751b6f4e70f2efaed85d6e0c0d89cefc44ac005a23c9877fc11bcea189cd5c
+size 4195636216

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "additional_special_tokens": [
+    "<|mol_smi_start|>",
+    "<|mol_smi_end|>",
+    "<|mol_sfi_start|>",
+    "<|mol_sfi_end|>",
+    "<|mol_sfi_h_start|>",
+    "<|mol_sfi_h_end|>",
+    "<|mol_3d_start|>",
+    "<|mol_3d_end|>",
+    "<|prot_aa_start|>",
+    "<|prot_aa_end|>",
+    "<|prot_3d_start|>",
+    "<|prot_3d_end|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:072d8c63700dabfae748c64dd88fdd218cd756a1220a19d60aa1277434fd7c73
+size 14441001