Upload folder using huggingface_hub

README.md

@@ -0,0 +1,111 @@
+---
+license: mit
+language:
+- en
+tags:
+- code
+- solidity
+- blockchain
+- smart-contracts
+- code-generation
+- ethereum
+- web3
+- daml
+base_model: Chain-GPT/Solidity-LLM
+pipeline_tag: text-generation
+library_name: transformers
+---
+<div align="center">
+# 🧞 Ginie — Solidity Smart Contract LLM
+**The AI built for Web3 developers**
+[![Website](https://img.shields.io/badge/Website-ginie.xyz-blue)](https://ginie.xyz)
+[![License](https://img.shields.io/badge/License-MIT-green)](https://opensource.org/licenses/MIT)
+[![Model](https://img.shields.io/badge/Model-2B_Parameters-orange)]()
+[![Solidity](https://img.shields.io/badge/Language-Solidity-purple)]()
+</div>
+---
+## What is Ginie?
+**Ginie AI** is the world's first specialized AI assistant for **Daml + Solidity smart contract generation**,
+built for the institutional blockchain ecosystem on [Canton Network](https://canton.network).
+This model — **Ginie Solidity LLM v1.0** — generates production-quality Solidity smart contracts
+from plain English instructions. It understands ERC standards, OpenZeppelin patterns, DeFi protocols,
+DAOs, governance, and more.
+> 🌐 Try the live demo: [huggingface.co/spaces/GinieAI/Ginie-Demo](https://huggingface.co/spaces/GinieAI/Ginie-Demo)
+---
+## Quickstart
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+tokenizer = AutoTokenizer.from_pretrained("GinieAI/Solidity-LLM")
+model = AutoModelForCausalLM.from_pretrained(
+    "GinieAI/Solidity-LLM",
+    torch_dtype=torch.bfloat16,
+    device_map="cuda"
+)
+prompt = """### Instruction:
+Write a Solidity ERC20 token contract with minting, burning, and owner controls.
+### Response:
+"""
+inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+outputs = model.generate(
+    **inputs,
+    max_new_tokens=800,
+    temperature=0.7,
+    do_sample=True,
+    pad_token_id=tokenizer.eos_token_id
+)
+print(tokenizer.decode(outputs, skip_special_tokens=True))
+```
+---
+## Model Details
+| Property | Value |
+|---|---|
+| **Developer** | [Ginie AI](https://ginie.xyz) |
+| **Model type** | Causal LM — Code Generation |
+| **Parameters** | 2 Billion |
+| **Architecture** | 32 Transformer blocks |
+| **Context length** | 2048 tokens |
+| **Precision** | bfloat16 |
+| **Tokenizer** | GPT-2 |
+| **Language** | Solidity (Ethereum-compatible) |
+| **License** | MIT |
+---
+## Performance Benchmarks
+| Metric | Ginie v1 | GPT-4o mini | DeepSeek-Coder-7B |
+|---|---|---|---|
+| Compilation Rate | **83%** | 78% | 75% |
+| OpenZeppelin Compliance | **65%** | 61% | 58% |
+| Gas Efficiency | **72%** | 65% | 63% |
+| Security Score | **58%** | 54% | 51% |
+---
+## Use Cases
+- ✅ Generating ERC20, ERC721, ERC1155 token contracts
+- ✅ DeFi protocols — staking, liquidity, yield farming
+- ✅ DAO and governance contracts
+- ✅ Multisig wallets and escrow
+- ✅ NFT marketplaces
+- ✅ Canton Network / Daml integration (v2 roadmap)
+## Not suitable for
+- ❌ Production deployment without expert review
+- ❌ Non-Solidity languages (v2 roadmap includes Rust/Daml)
+- ❌ Legal or formal contract auditing
+---
+## Roadmap
+| Version | What's new |
+|---|---|
+| **v1.0** (now) | Base Solidity generation — 2B params |
+| **v2.0** | +7,800 training examples — DISL + Zellic corpus |
+| **v3.0** | Daml smart contract support — Canton Network |
+| **v4.0** | Real user data flywheel — weekly retraining |
+---
+## License & Attribution
+This model is released under the **MIT License**.
+Built on top of [Chain-GPT/Solidity-LLM](https://huggingface.co/Chain-GPT/Solidity-LLM)
+by [ChainGPT](https://chaingpt.org), which is itself fine-tuned from
+[Salesforce/codegen-2B-multi](https://huggingface.co/Salesforce/codegen-2B-multi).
+Full credit to the original authors. Ginie AI extends this work for the
+institutional Canton/Daml ecosystem.
+---
+## About Ginie AI
+Ginie AI is building the developer layer for institutional blockchain.
+🌐 [ginie.xyz](https://ginie.xyz)

config.json

@@ -0,0 +1,52 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "CodeGenForCausalLM"
+  ],
+  "attn_pdrop": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "embd_pdrop": 0.0,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "codegen",
+  "n_ctx": 2048,
+  "n_embd": 2560,
+  "n_head": 32,
+  "n_inner": null,
+  "n_layer": 32,
+  "n_positions": 2048,
+  "resid_pdrop": 0.0,
+  "rotary_dim": 64,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50,
+      "temperature": 1.0
+    }
+  },
+  "tie_word_embeddings": false,
+  "tokenizer_class": "GPT2Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": true,
+  "vocab_size": 51200,
+  "_name_or_path": "GinieAI/Solidity-LLM",
+  "ginie_metadata": {
+    "model_name": "Ginie Solidity LLM",
+    "version": "1.0.0",
+    "developed_by": "Ginie AI",
+    "website": "https://ginie.xyz",
+    "base_model": "Chain-GPT/Solidity-LLM",
+    "original_base": "Salesforce/codegen-2B-multi",
+    "license": "MIT",
+    "use_case": "Solidity smart contract generation"
+  }
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 50256,
+  "transformers_version": "5.0.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b91b72289f6fb6deb0bcb39fedd445de67266f6181f31793fdd5a28d5252187
+size 5558740600

tokenizer_config.json

@@ -0,0 +1,13 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "is_local": false,
+  "model_max_length": 2048,
+  "model_specific_special_tokens": {},
+  "return_token_type_ids": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<|endoftext|>"
+}