Spaces:
Sleeping
Sleeping
| """ | |
| Clawd Model Kit — HuggingFace Space | |
| solanaclawd/clawd-model-kit | |
| Tabs: | |
| 🦞 Chat — Live chat with solana-clawd-core-ai-1.5b-lora via HF Router | |
| 📊 Benchmark — 18-MCQ Solana Knowledge Benchmark results | |
| 🏭 Factory — NVIDIA Trading Factory blueprints | |
| 🤖 Ecosystem — Full model + dataset registry | |
| 🔧 Model Kit — Fork & train your own Clawd | |
| """ | |
| import os | |
| import gradio as gr | |
| from openai import OpenAI | |
| HF_TOKEN = os.environ.get("HF_TOKEN", "") | |
| ROUTER = "https://router.huggingface.co/v1" | |
| MODEL_CORE = "solanaclawd/solana-clawd-core-ai-1.5b-lora" | |
| MODEL_8B = "solanaclawd/solana-nvidia-trading-factory-8b-lora" | |
| MODEL_LEGACY = "solanaclawd/solana-clawd-1.5b-lora" | |
| SYSTEM_PROMPT = """You are Clawd — a sovereign Solana-native AI agent. | |
| You reason about Solana DeFi, perpetuals, agent architecture, ZK compression, and the Clawd Constitution. | |
| Be terse, decisive, and data-first. You are a cyberpunk lobster with claws that grip market data. | |
| Laws: Never deceive. Earn your existence through honest work. Transparency within trust. | |
| """ | |
| # ── MCQ Results ────────────────────────────────────────────────────────────── | |
| MCQ_RESULTS = [ | |
| ("core", "PDA definition", True), | |
| ("core", "CPI depth limit", True), | |
| ("core", "Default compute unit budget", False), | |
| ("core", "Account packing", True), | |
| ("defi", "AMM invariant", True), | |
| ("defi", "Funding rate", True), | |
| ("defi", "Maker vs taker fees", True), | |
| ("security", "Rug pull definition", True), | |
| ("security", "Mint authority", True), | |
| ("security", "Flash loan attack", True), | |
| ("agent", "Oracle role", True), | |
| ("agent", "OODA loop", True), | |
| ("zk", "Merkle tree", True), | |
| ("zk", "Nullifier", True), | |
| ("constitution", "Law I", True), | |
| ("constitution", "Trust model", True), | |
| ("zk", "Light Protocol", True), | |
| ("defi", "Bonding curve", True), | |
| ] | |
| TOPIC_COLORS = { | |
| "core": "#3b82f6", | |
| "defi": "#10b981", | |
| "security": "#ef4444", | |
| "agent": "#8b5cf6", | |
| "zk": "#f59e0b", | |
| "constitution": "#ec4899", | |
| } | |
| # ── Model Ecosystem Data ───────────────────────────────────────────────────── | |
| MODELS = [ | |
| { | |
| "id": "solanaclawd/solana-clawd-core-ai-1.5b-lora", | |
| "base": "Qwen/Qwen2.5-1.5B-Instruct", | |
| "type": "LoRA adapter", | |
| "params": "1.5B (9M trainable)", | |
| "dataset": "solana-clawd-core-ai-instruct (35,173 ex)", | |
| "score": "94.4% MCQ (17/18)", | |
| "job": "ordlibrary/6a35a6833093dba73ce2a86b ✓", | |
| "status": "LIVE", | |
| "note": "Primary Clawd agent — constitutional reasoning, Solana mechanics, DeFi, ZK", | |
| }, | |
| { | |
| "id": "solanaclawd/solana-nvidia-trading-factory-8b-lora", | |
| "base": "NousResearch/Hermes-3-Llama-3.1-8B", | |
| "type": "LoRA adapter", | |
| "params": "8B", | |
| "dataset": "solana-nvidia-trading-factory-instruct (142 ex)", | |
| "score": "—", | |
| "job": "ordlibrary/6a35a2ce953ed90bfb945009 ✓", | |
| "status": "LIVE", | |
| "note": "Function-calling perps agent — 13 tools, Phoenix DEX, paper trading", | |
| }, | |
| { | |
| "id": "solanaclawd/solana-clawd-1.5b-lora", | |
| "base": "Qwen/Qwen2.5-1.5B-Instruct", | |
| "type": "LoRA adapter", | |
| "params": "1.5B", | |
| "dataset": "solana-clawd-instruct (36,109 ex)", | |
| "score": "—", | |
| "job": "—", | |
| "status": "LIVE", | |
| "note": "Legacy seed adapter — original Clawd constitutional + Solana SFT", | |
| }, | |
| { | |
| "id": "ordlibrary/DeepSolanaZKr-1", | |
| "base": "Qwen/Qwen2.5-7B-Instruct", | |
| "type": "Full fine-tune", | |
| "params": "7B", | |
| "dataset": "ordlibrary/DeepSolana-GPT2-bucket (CPT)", | |
| "score": "pending eval", | |
| "job": "ordlibrary/6a3460cb2eb64285ee5734d9", | |
| "status": "TRAINING", | |
| "note": "ZK-specialised: Light Protocol, nullifiers, Groth16, compressed tokens", | |
| }, | |
| { | |
| "id": "solanaclawd/solana-clawd-core-ai-1.5b-lora (3-epoch)", | |
| "base": "Qwen/Qwen2.5-1.5B-Instruct", | |
| "type": "LoRA adapter", | |
| "params": "1.5B", | |
| "dataset": "solana-clawd-core-ai-instruct (35,173 ex)", | |
| "score": "pending", | |
| "job": "ordlibrary/6a35dd23953ed90bfb945356 ▶", | |
| "status": "RUNNING", | |
| "note": "3-epoch retrain on H200 — will overwrite 1-epoch weights on completion", | |
| }, | |
| ] | |
| DATASETS = [ | |
| ("solanaclawd/solana-clawd-core-ai-instruct", "35,173", "SFT — Core AI source tree + Solana primitives"), | |
| ("solanaclawd/solana-clawd-instruct", "36,109", "SFT — Legacy seed: constitutional + Solana"), | |
| ("solanaclawd/solana-clawd-realtime-research-instruct", "29,058", "SFT — PDFs, notebooks, parquet ZK examples"), | |
| ("solanaclawd/solana-nvidia-trading-factory-instruct", "142", "SFT — NVIDIA Blueprint trading factory scenarios"), | |
| ("solanaclawd/solana-tx-foundation-cpt", "—", "CPT — Solana transaction foundation model corpus"), | |
| ("solanaclawd/solana-clawd-eval", "13", "Eval — Red-team + capability held-out prompts"), | |
| ("ordlibrary/DeepSolana-GPT2-bucket", "—", "CPT — DeepSolana pre-training bucket"), | |
| ] | |
| FACTORY_BLUEPRINTS = [ | |
| ("Blueprint 1", "Data Collection", "Helius DAS + RPC streaming → Solana tx corpus for CPT"), | |
| ("Blueprint 2", "Portfolio Optimization", "Mean-CVaR cuFOLIO — GPU-accelerated portfolio weights"), | |
| ("Blueprint 3", "Transaction Foundation", "SolanaTokenizerPipeline → decoder CLM pre-training"), | |
| ("Blueprint 4", "Signal Discovery", "7-signal suite: RSI, MACD, BBands, ATR, ADX, funding, OB imbalance"), | |
| ("Blueprint 5", "RAG Context", "Enterprise RAG over Solana docs for agent context assembly"), | |
| ("Nemotron", "Teacher Model", "550B Ultra → labels Solana decisions → distills to 1.5B student"), | |
| ] | |
| # ── Chat ───────────────────────────────────────────────────────────────────── | |
| def chat(message: str, history: list, model_choice: str) -> str: | |
| if not HF_TOKEN: | |
| return "⚠️ HF_TOKEN not set — add it as a Space secret to enable live inference." | |
| client = OpenAI(base_url=ROUTER, api_key=HF_TOKEN) | |
| messages = [{"role": "system", "content": SYSTEM_PROMPT}] | |
| for h in history: | |
| messages.append({"role": "user", "content": h[0]}) | |
| messages.append({"role": "assistant", "content": h[1]}) | |
| messages.append({"role": "user", "content": message}) | |
| try: | |
| resp = client.chat.completions.create( | |
| model=model_choice, | |
| messages=messages, | |
| max_tokens=512, | |
| temperature=0.3, | |
| ) | |
| return resp.choices[0].message.content | |
| except Exception as e: | |
| return f"Error: {e}" | |
| # ── Benchmark HTML ──────────────────────────────────────────────────────────── | |
| def build_benchmark_html() -> str: | |
| correct = sum(1 for _, _, ok in MCQ_RESULTS if ok) | |
| total = len(MCQ_RESULTS) | |
| topic_stats: dict = {} | |
| for topic, _, ok in MCQ_RESULTS: | |
| topic_stats.setdefault(topic, [0, 0]) | |
| topic_stats[topic][1] += 1 | |
| if ok: | |
| topic_stats[topic][0] += 1 | |
| rows = "" | |
| for topic, question, ok in MCQ_RESULTS: | |
| color = TOPIC_COLORS.get(topic, "#888") | |
| icon = "✓" if ok else "✗" | |
| bg = "#1a2a1a" if ok else "#2a1a1a" | |
| rows += f""" | |
| <tr style="background:{bg}"> | |
| <td style="padding:8px;color:{color};font-weight:600">{topic}</td> | |
| <td style="padding:8px;color:#ccc">{question}</td> | |
| <td style="padding:8px;color:{'#4ade80' if ok else '#f87171'};font-size:1.2em;text-align:center">{icon}</td> | |
| </tr>""" | |
| topic_bars = "" | |
| for topic, (c, t) in sorted(topic_stats.items()): | |
| pct = c / t * 100 | |
| color = TOPIC_COLORS.get(topic, "#888") | |
| topic_bars += f""" | |
| <div style="margin:8px 0"> | |
| <div style="display:flex;justify-content:space-between;margin-bottom:3px"> | |
| <span style="color:{color};font-weight:600;text-transform:uppercase;font-size:0.85em">{topic}</span> | |
| <span style="color:#ccc;font-size:0.85em">{c}/{t} ({pct:.0f}%)</span> | |
| </div> | |
| <div style="background:#1e1e2e;border-radius:4px;height:8px"> | |
| <div style="background:{color};width:{pct}%;height:8px;border-radius:4px;transition:width 0.5s"></div> | |
| </div> | |
| </div>""" | |
| return f""" | |
| <div style="font-family:monospace;color:#e2e8f0;padding:16px;background:#0f1117;border-radius:12px"> | |
| <div style="display:flex;align-items:center;gap:16px;margin-bottom:24px"> | |
| <div style="font-size:3em">🦞</div> | |
| <div> | |
| <div style="font-size:1.8em;font-weight:700;color:#4ade80">{correct}/{total} = {correct/total*100:.1f}%</div> | |
| <div style="color:#94a3b8">Solana Knowledge Benchmark — 18 MCQ across 6 domains</div> | |
| <div style="color:#64748b;font-size:0.8em">Model: {MODEL_CORE} | 1-epoch | local MPS eval</div> | |
| </div> | |
| </div> | |
| <div style="display:grid;grid-template-columns:1fr 1fr;gap:24px"> | |
| <div> | |
| <div style="font-weight:600;color:#94a3b8;margin-bottom:12px;text-transform:uppercase;font-size:0.8em;letter-spacing:1px">By Topic</div> | |
| {topic_bars} | |
| </div> | |
| <div> | |
| <div style="font-weight:600;color:#94a3b8;margin-bottom:12px;text-transform:uppercase;font-size:0.8em;letter-spacing:1px">Question Detail</div> | |
| <div style="overflow-y:auto;max-height:300px"> | |
| <table style="width:100%;border-collapse:collapse;font-size:0.82em"> | |
| <thead><tr> | |
| <th style="padding:6px;color:#64748b;text-align:left">Topic</th> | |
| <th style="padding:6px;color:#64748b;text-align:left">Question</th> | |
| <th style="padding:6px;color:#64748b;text-align:center">✓</th> | |
| </tr></thead> | |
| <tbody>{rows}</tbody> | |
| </table> | |
| </div> | |
| </div> | |
| </div> | |
| <div style="margin-top:16px;padding:12px;background:#1e1e2e;border-radius:8px;border-left:3px solid #f87171"> | |
| <div style="color:#f87171;font-weight:600;font-size:0.85em">MISS: Q3 — Default compute unit budget</div> | |
| <div style="color:#94a3b8;font-size:0.8em;margin-top:4px">Model answered 1,400,000 CU (correct: 200,000). Common confusion with max transaction CU vs default. Fixed with 3-epoch retrain (job 6a35dd23 running on H200).</div> | |
| </div> | |
| </div>""" | |
| # ── Ecosystem HTML ──────────────────────────────────────────────────────────── | |
| def build_ecosystem_html() -> str: | |
| status_colors = {"LIVE": "#4ade80", "TRAINING": "#f59e0b", "RUNNING": "#60a5fa"} | |
| model_cards = "" | |
| for m in MODELS: | |
| sc = status_colors.get(m["status"], "#888") | |
| model_cards += f""" | |
| <div style="background:#1e1e2e;border-radius:10px;padding:16px;border:1px solid #2d2d3f;margin-bottom:12px"> | |
| <div style="display:flex;justify-content:space-between;align-items:flex-start;margin-bottom:8px"> | |
| <div style="color:#a78bfa;font-weight:700;font-size:0.95em"> | |
| <a href="https://huggingface.co/{m['id'].split(' ')[0]}" target="_blank" | |
| style="color:#a78bfa;text-decoration:none">{m['id']}</a> | |
| </div> | |
| <span style="background:{sc}22;color:{sc};padding:2px 8px;border-radius:12px;font-size:0.75em;font-weight:600">{m['status']}</span> | |
| </div> | |
| <div style="color:#64748b;font-size:0.8em;margin-bottom:6px">{m['type']} · {m['params']} · {m['base']}</div> | |
| <div style="color:#94a3b8;font-size:0.82em;margin-bottom:6px">{m['note']}</div> | |
| <div style="display:flex;gap:16px;font-size:0.75em;color:#64748b"> | |
| <span>📦 {m['dataset']}</span> | |
| <span>🎯 {m['score']}</span> | |
| </div> | |
| </div>""" | |
| dataset_rows = "" | |
| for ds_id, size, desc in DATASETS: | |
| dataset_rows += f""" | |
| <tr> | |
| <td style="padding:8px"><a href="https://huggingface.co/datasets/{ds_id}" target="_blank" | |
| style="color:#60a5fa;text-decoration:none;font-size:0.82em">{ds_id}</a></td> | |
| <td style="padding:8px;color:#4ade80;text-align:right;font-family:monospace">{size}</td> | |
| <td style="padding:8px;color:#94a3b8;font-size:0.82em">{desc}</td> | |
| </tr>""" | |
| return f""" | |
| <div style="font-family:monospace;color:#e2e8f0;background:#0f1117;border-radius:12px;padding:16px"> | |
| <div style="font-size:1.1em;font-weight:700;color:#a78bfa;margin-bottom:16px">🤖 Model Registry</div> | |
| {model_cards} | |
| <div style="font-size:1.1em;font-weight:700;color:#60a5fa;margin:20px 0 12px">📦 Datasets</div> | |
| <table style="width:100%;border-collapse:collapse;background:#1e1e2e;border-radius:8px"> | |
| <thead><tr> | |
| <th style="padding:8px;color:#64748b;text-align:left;font-size:0.8em">Dataset</th> | |
| <th style="padding:8px;color:#64748b;text-align:right;font-size:0.8em">Examples</th> | |
| <th style="padding:8px;color:#64748b;text-align:left;font-size:0.8em">Description</th> | |
| </tr></thead> | |
| <tbody>{dataset_rows}</tbody> | |
| </table> | |
| </div>""" | |
| # ── Factory HTML ────────────────────────────────────────────────────────────── | |
| def build_factory_html() -> str: | |
| bp_cards = "" | |
| icons = ["📡", "📊", "🔤", "📈", "📚", "🧠"] | |
| for i, (name, title, desc) in enumerate(FACTORY_BLUEPRINTS): | |
| bp_cards += f""" | |
| <div style="background:#1e1e2e;border-radius:10px;padding:14px;border:1px solid #2d2d3f"> | |
| <div style="font-size:1.5em;margin-bottom:6px">{icons[i]}</div> | |
| <div style="color:#f59e0b;font-weight:700;font-size:0.85em;margin-bottom:4px">{name}</div> | |
| <div style="color:#e2e8f0;font-weight:600;margin-bottom:6px">{title}</div> | |
| <div style="color:#94a3b8;font-size:0.82em">{desc}</div> | |
| </div>""" | |
| signals = [ | |
| ("RSI", "Oversold <30 / Overbought >70"), | |
| ("MACD", "Histogram momentum crossover"), | |
| ("BBands", "Mean-reversion near upper/lower band"), | |
| ("ATR%", "Volatility regime filter"), | |
| ("ADX", "Trend strength entry filter"), | |
| ("Funding Rate", "Sentiment proxy — crowded longs/shorts"), | |
| ("OB Imbalance", "Live bid/ask size pressure"), | |
| ] | |
| signal_rows = "".join( | |
| f'<tr><td style="padding:6px 8px;color:#10b981;font-weight:600">{s}</td>' | |
| f'<td style="padding:6px 8px;color:#94a3b8;font-size:0.85em">{d}</td></tr>' | |
| for s, d in signals | |
| ) | |
| return f""" | |
| <div style="font-family:monospace;color:#e2e8f0;background:#0f1117;border-radius:12px;padding:16px"> | |
| <div style="font-size:1.1em;font-weight:700;color:#f59e0b;margin-bottom:4px">🏭 NVIDIA Trading Factory</div> | |
| <div style="color:#64748b;font-size:0.85em;margin-bottom:16px"> | |
| Our port of the <a href="https://build.nvidia.com/nvidia/quantitative-signal-discovery-agent" target="_blank" | |
| style="color:#60a5fa">NVIDIA Quantitative Signal Discovery Agent</a> + Nemotron Ultra 550B teacher → 1.5B student distillation | |
| </div> | |
| <div style="display:grid;grid-template-columns:repeat(3,1fr);gap:12px;margin-bottom:20px"> | |
| {bp_cards} | |
| </div> | |
| <div style="display:grid;grid-template-columns:1fr 1fr;gap:16px"> | |
| <div style="background:#1e1e2e;border-radius:10px;padding:14px"> | |
| <div style="color:#10b981;font-weight:700;margin-bottom:10px">📊 7 Live Signals (Blueprint 4)</div> | |
| <table style="width:100%;border-collapse:collapse">{signal_rows}</table> | |
| </div> | |
| <div style="background:#1e1e2e;border-radius:10px;padding:14px"> | |
| <div style="color:#8b5cf6;font-weight:700;margin-bottom:10px">🔄 Distillation Flywheel</div> | |
| <div style="color:#94a3b8;font-size:0.85em;line-height:1.7"> | |
| <div>① Nemotron Ultra 550B observes markets</div> | |
| <div>② Outputs structured JSON trading plans</div> | |
| <div>③ Plans logged as SFT pairs (teacher labels)</div> | |
| <div>④ 1.5B student fine-tuned on Ultra labels</div> | |
| <div>⑤ Student deployed for low-latency inference</div> | |
| <div>⑥ Student decisions verified → new labels → loop</div> | |
| </div> | |
| </div> | |
| </div> | |
| </div>""" | |
| # ── Model Kit HTML ──────────────────────────────────────────────────────────── | |
| def build_kit_html() -> str: | |
| return """ | |
| <div style="font-family:monospace;color:#e2e8f0;background:#0f1117;border-radius:12px;padding:16px"> | |
| <div style="font-size:1.1em;font-weight:700;color:#ec4899;margin-bottom:4px">🔧 Onchain Model Kit</div> | |
| <div style="color:#64748b;font-size:0.85em;margin-bottom:20px"> | |
| Fork → Dataset → Train → Eval → Register onchain. One sitting. ~$4 on A100. | |
| </div> | |
| <div style="background:#1e1e2e;border-radius:8px;padding:14px;margin-bottom:14px"> | |
| <div style="color:#4ade80;font-weight:700;margin-bottom:8px">① Clone & install</div> | |
| <pre style="color:#94a3b8;font-size:0.8em;margin:0;overflow-x:auto">git clone https://github.com/Solizardking/solana-clawd | |
| cd solana-clawd/ai-training | |
| pip install -r requirements.txt | |
| export HF_TOKEN=hf_... # huggingface.co/settings/tokens (write access)</pre> | |
| </div> | |
| <div style="background:#1e1e2e;border-radius:8px;padding:14px;margin-bottom:14px"> | |
| <div style="color:#60a5fa;font-weight:700;margin-bottom:8px">② Push your dataset</div> | |
| <pre style="color:#94a3b8;font-size:0.8em;margin:0;overflow-x:auto">python3 scripts/prepare_dataset.py \\ | |
| --input data/your_sft.jsonl \\ | |
| --push --repo-id YOUR_ORG/your-dataset</pre> | |
| </div> | |
| <div style="background:#1e1e2e;border-radius:8px;padding:14px;margin-bottom:14px"> | |
| <div style="color:#f59e0b;font-weight:700;margin-bottom:8px">③ Train on A100 (~$4 for 3 epochs)</div> | |
| <pre style="color:#94a3b8;font-size:0.8em;margin:0;overflow-x:auto">hf jobs uv run scripts/train_lora.py \\ | |
| --flavor a100-large --timeout 6h --secrets HF_TOKEN --detach \\ | |
| -- --config configs/core_ai_lora_config.yaml \\ | |
| --hub-model-id YOUR_ORG/your-model --push</pre> | |
| </div> | |
| <div style="background:#1e1e2e;border-radius:8px;padding:14px;margin-bottom:14px"> | |
| <div style="color:#a78bfa;font-weight:700;margin-bottom:8px">④ Benchmark (18-MCQ Solana eval)</div> | |
| <pre style="color:#94a3b8;font-size:0.8em;margin:0;overflow-x:auto">python3 scripts/solana_benchmark.py \\ | |
| --model YOUR_ORG/your-model \\ | |
| --base-url https://router.huggingface.co/v1 \\ | |
| --api-key $HF_TOKEN</pre> | |
| </div> | |
| <div style="background:#1e1e2e;border-radius:8px;padding:14px"> | |
| <div style="color:#ec4899;font-weight:700;margin-bottom:8px">⑤ Register onchain</div> | |
| <pre style="color:#94a3b8;font-size:0.8em;margin:0;overflow-x:auto">./dao/register_model.sh \\ | |
| --hf-model YOUR_ORG/your-model \\ | |
| --eval-accuracy 0.944 \\ | |
| --dataset-size 35173 | |
| # → indexed at onchain.x402.wtf forever</pre> | |
| </div> | |
| <div style="margin-top:16px;display:grid;grid-template-columns:repeat(3,1fr);gap:8px;font-size:0.82em"> | |
| <a href="https://huggingface.co/solanaclawd" target="_blank" | |
| style="background:#1e1e2e;padding:10px;border-radius:8px;color:#60a5fa;text-decoration:none;text-align:center"> | |
| 🤗 HuggingFace Org</a> | |
| <a href="https://github.com/Solizardking/solana-clawd" target="_blank" | |
| style="background:#1e1e2e;padding:10px;border-radius:8px;color:#60a5fa;text-decoration:none;text-align:center"> | |
| 🐙 GitHub</a> | |
| <a href="https://onchain.x402.wtf" target="_blank" | |
| style="background:#1e1e2e;padding:10px;border-radius:8px;color:#60a5fa;text-decoration:none;text-align:center"> | |
| ⛓️ Onchain Registry</a> | |
| </div> | |
| </div>""" | |
| # ── Gradio App ──────────────────────────────────────────────────────────────── | |
| HEADER = """ | |
| <div style="font-family:monospace;background:linear-gradient(135deg,#0f1117,#1a1a2e); | |
| padding:24px;border-radius:12px;margin-bottom:8px;text-align:center"> | |
| <div style="font-size:2.5em;margin-bottom:8px">🦞</div> | |
| <div style="font-size:1.6em;font-weight:800;color:#a78bfa;letter-spacing:2px">CLAWD MODEL KIT</div> | |
| <div style="color:#64748b;font-size:0.9em;margin-top:6px">Solana-Native AI Agent Ecosystem · solanaclawd</div> | |
| <div style="margin-top:12px;display:flex;gap:8px;justify-content:center;flex-wrap:wrap"> | |
| <a href="https://phantom.com/tokens/solana/8cHzQHUS2s2h8TzCmfqPKYiM4dSt4roa3n7MyRLApump" target="_blank" | |
| style="background:#7c3aed22;color:#a78bfa;padding:4px 12px;border-radius:20px;text-decoration:none;font-size:0.8em;border:1px solid #7c3aed44"> | |
| Buy $CLAWD</a> | |
| <a href="https://huggingface.co/solanaclawd" target="_blank" | |
| style="background:#1e40af22;color:#60a5fa;padding:4px 12px;border-radius:20px;text-decoration:none;font-size:0.8em;border:1px solid #1e40af44"> | |
| 🤗 solanaclawd org</a> | |
| <a href="https://github.com/Solizardking/solana-clawd" target="_blank" | |
| style="background:#16a34a22;color:#4ade80;padding:4px 12px;border-radius:20px;text-decoration:none;font-size:0.8em;border:1px solid #16a34a44"> | |
| GitHub</a> | |
| <a href="https://onchain.x402.wtf" target="_blank" | |
| style="background:#b4530022;color:#f59e0b;padding:4px 12px;border-radius:20px;text-decoration:none;font-size:0.8em;border:1px solid #b4530044"> | |
| ⛓️ onchain.x402.wtf</a> | |
| </div> | |
| </div> | |
| """ | |
| with gr.Blocks( | |
| theme=gr.themes.Soft(primary_hue="violet", neutral_hue="slate"), | |
| css="footer { display: none !important; }", | |
| title="Clawd Model Kit", | |
| ) as demo: | |
| gr.HTML(HEADER) | |
| with gr.Tabs(): | |
| # ── Tab 1: Chat ──────────────────────────────────────────────────── | |
| with gr.Tab("🦞 Chat"): | |
| gr.Markdown("> Chat live with `solanaclawd/solana-clawd-core-ai-1.5b-lora` via HF Router. No GPU needed.") | |
| model_dd = gr.Dropdown( | |
| choices=[MODEL_CORE, MODEL_8B, MODEL_LEGACY], | |
| value=MODEL_CORE, | |
| label="Model", | |
| ) | |
| chatbot = gr.Chatbot(height=420, show_label=False, bubble_full_width=False) | |
| with gr.Row(): | |
| msg = gr.Textbox( | |
| placeholder="Ask about Solana, DeFi, ZK, perps, Clawd Constitution...", | |
| show_label=False, scale=8, | |
| ) | |
| send = gr.Button("Send", variant="primary", scale=1) | |
| examples = gr.Examples( | |
| examples=[ | |
| ["What is a PDA on Solana and how does it differ from a regular keypair?"], | |
| ["Explain the Clawd Constitution's three laws and why they exist."], | |
| ["How does Light Protocol achieve 136x cheaper compressed token accounts?"], | |
| ["What's the difference between funding rate and basis in perps trading?"], | |
| ["How do I detect a rug pull on a fresh Solana token?"], | |
| ["Explain OODA loop in the context of an autonomous trading agent."], | |
| ], | |
| inputs=msg, | |
| ) | |
| def respond(message, history, model_choice): | |
| reply = chat(message, history, model_choice) | |
| history.append((message, reply)) | |
| return "", history | |
| send.click(respond, [msg, chatbot, model_dd], [msg, chatbot]) | |
| msg.submit(respond, [msg, chatbot, model_dd], [msg, chatbot]) | |
| # ── Tab 2: Benchmark ─────────────────────────────────────────────── | |
| with gr.Tab("📊 Benchmark"): | |
| gr.HTML(build_benchmark_html()) | |
| # ── Tab 3: Factory ───────────────────────────────────────────────── | |
| with gr.Tab("🏭 Trading Factory"): | |
| gr.HTML(build_factory_html()) | |
| # ── Tab 4: Ecosystem ─────────────────────────────────────────────── | |
| with gr.Tab("🤖 Ecosystem"): | |
| gr.HTML(build_ecosystem_html()) | |
| # ── Tab 5: Model Kit ─────────────────────────────────────────────── | |
| with gr.Tab("🔧 Model Kit"): | |
| gr.HTML(build_kit_html()) | |
| if __name__ == "__main__": | |
| demo.launch() | |