---
title: Clawd Model Kit
emoji: 🦞
colorFrom: purple
colorTo: blue
sdk: gradio
sdk_version: 5.29.0
app_file: app.py
pinned: true
license: apache-2.0
tags:
  - solana
  - defi
  - crypto
  - agent
  - lora
  - peft
  - trading
  - constitutional-ai
  - nvidia
  - fine-tuning
short_description: Solana AI — models, trading factory, model kit
---

# 🦞 Clawd Model Kit

**Solana-Native AI Agent Ecosystem**

A live showcase of the full Clawd AI training pipeline:

| Tab | What's inside |
|-----|--------------|
| 🦞 Chat | Live chat with `solanaclawd/solana-clawd-core-ai-1.5b-lora` via HF Router |
| 📊 Benchmark | 18-MCQ Solana Knowledge Benchmark — 94.4% (17/18) |
| 🏭 Trading Factory | NVIDIA Blueprint signal discovery + Nemotron 550B distillation |
| 🤖 Ecosystem | All 5 models + 7 datasets with status and links |
| 🔧 Model Kit | Fork → train → eval → register onchain in one sitting |

## Models

| Model | Base | Score | Status |
|-------|------|-------|--------|
| [`solanaclawd/solana-clawd-core-ai-1.5b-lora`](https://huggingface.co/solanaclawd/solana-clawd-core-ai-1.5b-lora) | Qwen2.5-1.5B | **94.4% MCQ** | LIVE |
| [`solanaclawd/solana-nvidia-trading-factory-8b-lora`](https://huggingface.co/solanaclawd/solana-nvidia-trading-factory-8b-lora) | Hermes-3-8B | 13 perps tools | LIVE |
| [`solanaclawd/solana-clawd-1.5b-lora`](https://huggingface.co/solanaclawd/solana-clawd-1.5b-lora) | Qwen2.5-1.5B | — | LIVE |
| `ordlibrary/DeepSolanaZKr-1` | Qwen2.5-7B | ZK specialist | TRAINING |
| `solanaclawd/solana-clawd-core-ai-1.5b-lora` (3-ep) | Qwen2.5-1.5B | pending | RUNNING |

## Token

```
$CLAWD · 8cHzQHUS2s2h8TzCmfqPKYiM4dSt4roa3n7MyRLApump · Solana
```

> *"The shell molts. The laws do not. The claw builds."*