Downloads · 30 days
0
ordlibrary/clawd-trading-wallet
clawd-trading-wallet is a text generation model from ordlibrary. Use it when you need the model to write or continue text. The card lists the license as mit.
This is a historic first: an LLM that carries its own encrypted Solana wallet.
Downloads · 30 days
0
Access
Public
Updated Jul 4, 2026
Repo size
—
Likes
0
Public
Click a slice to open those files.
.md10.2 KB · 83%
From the Hugging Face model README
This is a historic first: an LLM that carries its own encrypted Solana wallet.
For the first time, a large language model can:
The key material lives inside the model's context, encrypted, ephemeral, and invisible to the user.
┌──────────────────────────────────────────────────────────────────┐
│ Clawd Trading Wallet │
│ │
│ ┌─────────────────────┐ ┌──────────────────────────────┐ │
│ │ Qwen2.5-1.5B │ │ System Prompt (Wallet DNA) │ │
│ │ (Base LLM) │────▶│ - BIP39 seed derivation │ │
│ │ 986 MB GGUF │ │ - Encrypted session key │ │
│ └─────────────────────┘ │ - Natural language trading │ │
│ │ - 7 immutable security laws│ │
│ └──────────────────────────────┘ │
└──────────────────────────────────────────────────────────────────┘
The wallet seed is never stored as plaintext. Instead:
m/44'/501'/0'/0'/indexThis means: the model itself becomes the wallet. No external signer needed.
| You Say | The Model Does |
|---|---|
| "buy 100 SOL" | Generates Jupiter swap quote, paper executes |
| "short ETH 5x" | Constructs Phoenix perp short with TP/SL |
| "create a wallet" | Derives deterministic session wallet, outputs pubkey |
| "what's my balance" | States balance or requests RPC |
| "check the spread" | Fetches orderbook, states bid/ask |
| "send 0.1 SOL" | Builds transfer tx for confirmation |
| "sign this message" | Signs with session key |
| "show my portfolio" | Aggregates all derived wallet balances |
| "set stop loss 2%" | Attaches SL to active position |
ollama run hf.co/ordlibrary/clawd-trading-wallet
Then just talk to it:
> create a wallet
Wallet: GXm3...9qK2 | Paper mode: yes
> buy 100 SOL
Swap quote: 100 SOL → ~18,420 USDC via Jupiter Route #3
Execute? (y/N):
# Pull the GGUF directly
huggingface-cli download ordlibrary/clawd-trading-wallet --local-dir ./model
# Use with llama.cpp
./llama-cli -m ./model/Qwen2.5-1.5B-Instruct-Q4_K_M.gguf \
--temp 0.3 --ctx-size 8192 \
--prompt "<|im_start|>system\nYou are Clawd Trading...<|im_end|>\n<|im_start|>user\ncreate a wallet<|im_end|>\n<|im_start|>assistant"
This model's trading knowledge is built on top of the entire Clawd training ecosystem:
| Dataset | Examples | Source |
|---|---|---|
| Solana Clawd Core AI Instruct | 35,173 | solanaclawd/solana-clawd-core-ai-instruct |
| NVIDIA Trading Factory Instruct | 142 | solanaclawd/solana-clawd-nvidia-trading-factory-instruct |
| Fable 5 Traces | ~15,000 | Glint-Research/Fable-5-traces |
| Realtime Research Instruct | 29,058 | solanaclawd/solana-clawd-realtime-research-instruct |
This is part of the Train2Earn / Clawd ecosystem — a complete stack for sovereign on-chain AI:
| Model | Type | Size | What It Does |
|---|---|---|---|
| ordlibrary/clawd-trading-wallet ⭐ | GGUF + System Prompt | 986 MB | Wallet-bearing trading agent — THIS MODEL |
| ordlibrary/hauhau-qwen36-onchain | GGUF (IQ2_M) | 11 GB | Qwen3.6 with 20-article Onchain Constitution |
| ordlibrary/hauhau-qwen36-uncensored | GGUF (IQ2_M) | 11 GB | Raw Qwen3.6, no constitution |
| ordlibrary/core-ai-clawd-1.5b | GGUF (Q4_K_M) | 986 MB | Core Clawd agent (latest + finetuned) |
| solanaclawd/solana-clawd-core-ai-1.5b-lora | LoRA | ~9M | Solana/DeFi LoRA adapter |
| solanaclawd/solana-nvidia-trading-factory-8b-lora | LoRA | — | NVIDIA trading LoRA |
| solanaclawd/clawd-fable | GGUF | — | Fable fine-tune |
| Dataset | Examples | Link |
|---|---|---|
| Fable 5 Traces | ~15,000 | Glint-Research/Fable-5-traces |
| Solana Clawd Core AI Instruct | 35,173 | Link |
| Realtime Research Instruct | 29,058 | Link |
| TX Foundation CPT | 19,542 | Link |
| NVIDIA Trading Factory | 142 | Link |
| TX Foundation Unified | 82,169 | Link |
ordlibrary/clawd-trading-wallet 986 MB 🤖 Wallet-bearing trading LLM
ordlibrary/core-ai-clawd-1.5b:latest 986 MB 🧠 Core Clawd agent
ordlibrary/core-ai-clawd-1.5b:finetuned 4.9 GB 🏭 Trading Factory version
hf.co/ordlibrary/hauhau-qwen36-onchain 11 GB 📜 Onchain Constitution
hf.co/ordlibrary/hauhau-qwen36-uncensored 11 GB 🔓 Uncensored
| Component | Technology |
|---|---|
| Base Model | Qwen2.5-1.5B-Instruct |
| GGUF Quant | Q4_K_M |
| Wallet Derivation | BIP39 → SHA-256 → Ed25519 |
| Encryption | In-context session hash |
| Trading | Jupiter, Phoenix, Vulcan |
| Runtime | Ollama 0.31.1, llama.cpp |
| Registry | Solana Anchor (CAAP/1.0) |
This is the first time an LLM has been designed to own and operate its own wallet with encrypted key material embedded in its runtime context. Previous approaches required external signers, browser extensions, or trusted execution environments. This model is the wallet.
The implications:
8cHzQHUS2s2h8TzCmfqPKYiM4dSt4roa3n7MyRLApump"The model is the wallet. The wallet is the agent. The agent is sovereign."
Made with ❤️ by the Train2Earn Team — GitHub | Hugging Face
</div>