| name | consult-cloudflaremodels |
| description | Query multiple AI models via Cloudflare Workers AI API for second opinions on trading data, code, or analysis. Use when the user says "/consult-cloudflaremodels", "ask cloudflare models", or wants peer model consensus via Cloudflare's AI gateway. |
consult-cloudflaremodels
Query multiple Cloudflare Workers AI models for consensus analysis, non-interactively.
When to use
User asks "ask cloudflare models" / "/consult-cloudflaremodels" / multi-model second opinion from Cloudflare Workers AI.
Prerequisites
- Cloudflare Account ID and API Token (stored in environment or provided by user)
- API endpoint:
https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/run/{MODEL}
- Key models available:
@cf/meta/llama-4-scout-17b-16e-instruct, @cf/deepseek-ai/deepseek-r1-distill-qwen-32b, @cf/qwen/qwen3-235b-a22b, @cf/meta/llama-3.1-405b-instruct
The one command that works
python3 -c "
import json, subprocess, time
ACCOUNT_ID = 'YOUR_CLOUDFLARE_ACCOUNT_ID'
API_TOKEN = 'YOUR_CLOUDFLARE_API_TOKEN'
BASE_URL = f'https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/run'
PROMPT = 'YOUR_ANALYSIS_PROMPT_HERE'
models = [
'@cf/meta/llama-4-scout-17b-16e-instruct',
'@cf/deepseek-ai/deepseek-r1-distill-qwen-32b',
'@cf/qwen/qwen3-235b-a22b',
'@cf/meta/llama-3.1-405b-instruct',
]
for model in models:
payload = {
'messages': [{'role': 'user', 'content': PROMPT}],
'max_tokens': 1500,
'temperature': 0.3,
}
try:
resp = subprocess.run([
'curl', '-s', '--max-time', '90',
f'{BASE_URL}/{model}',
'-H', f'Authorization: Bearer {API_TOKEN}',
'-H', 'Content-Type: application/json',
'-d', json.dumps(payload)
], capture_output=True, text=True, timeout=95)
data = json.loads(resp.stdout)
if data.get('success'):
content = data['result']['response']
print(f'=== {model} ===')
print(content[:2000])
else:
print(f'=== {model} === ERROR: {data.get(\"errors\",[{}])[0].get(\"message\",\"unknown\")}')
except Exception as e:
print(f'=== {model} === EXCEPTION: {e}')
time.sleep(0.5)
"
Recommended Models (Cloudflare Workers AI)
| Model | Provider | Notes |
|---|
@cf/meta/llama-4-scout-17b-16e-instruct | Meta | Top performer in our tournament (WR=61.4%, Sharpe=0.326) |
@cf/deepseek-ai/deepseek-r1-distill-qwen-32b | DeepSeek/Qwen | Strong reasoning, good for coding/math |
@cf/qwen/qwen3-235b-a22b | Alibaba | Multilingual, good instruction following |
@cf/meta/llama-3.3-70b-instruct | Meta | Reliable daily driver |
@cf/meta/llama-3.2-3b-instruct | Meta | Fast/cheap for quick checks |
Known issues
- Account setup required — Cloudflare Workers AI requires account setup and API token generation at dash.cloudflare.com
- Rate limits — Free tier has daily request limits; add sleep between calls
- Model availability — Not all models are available in all regions; check Cloudflare docs for current catalog
- Response format — Cloudflare wraps responses differently than OpenAI API; parse
result.response field
Related skills
consult-nvidiamodels — Multi-model consensus via NVIDIA NIM API (primary recommended skill)consult-gemini — Single-model second opinion from Google Gemini
