AI access before you're ready to pay.
Your attention has value. Turn verified developer spotlights into promotional AI credits, or purchase persistent cloud packs when you're ready. Spend seamlessly via standard OpenAI client libraries.
curl https://api.attenval.com/v1/chat/completions \
-H "Authorization: Bearer atv_live_9f8a3d12bc784a9" \
-H "Content-Type: application/json" \
-d '{
"model": "attenval-auto",
"messages": [
{ "role": "system", "content": "You are a concise systems assistant." },
{ "role": "user", "content": "How does Attenval route prompts dynamically?" }
]
}' from openai import OpenAI
client = OpenAI(
api_key="atv_live_9f8a3d12bc784a9",
base_url="https://api.attenval.com/v1"
)
response = client.chat.completions.create(
model="attenval-auto",
messages=[
{"role": "system", "content": "You are a concise systems assistant."},
{"role": "user", "content": "How does Attenval route prompts dynamically?"}
]
)
print(response.choices[0].message.content)
# Access custom telemetry
print(f"Credits used: {response.attenval.credits_used}") import OpenAI from 'openai';
const client = new OpenAI({
apiKey: 'atv_live_9f8a3d12bc784a9',
baseURL: 'https://api.attenval.com/v1',
});
const completion = await client.chat.completions.create({
model: 'attenval-auto',
messages: [
{ role: 'system', content: 'You are a concise systems assistant.' },
{ role: 'user', content: 'How does Attenval route prompts dynamically?' },
],
});
console.log(completion.choices[0]?.message.content); How Attenval Works
A closed-loop pipeline transforming verified sponsor attention into cryptographic ledger credits and low-latency model inference.
Verified Attention
Engage with verified developer spotlights in the mobile app. Cryptographic server-side verification (SSV) validates completion before granting credits.
Durable DO Ledger
Each account operates a dedicated SQLite Durable Object. Concurrency-safe two-phase reserve and settle protocol prevents negative balances.
Intelligent AI Routing
Global Cloudflare AI Gateway routes requests to Workers AI edge nodes, DeepSeek, or specialized coding engines with automatic failover.
One API Key, Four Specialized Aliases
Never manage individual upstream keys again. Target clean aliases optimized for latency, code intelligence, and deep reasoning.
attenval-auto
RecommendedDeepSeek V3, Workers AI, Llama 3.3 Analyzes prompt structure and routes to the most cost-effective tier with optimal response depth.
attenval-fast
Low LatencyWorkers AI Edge, Llama 3.1 8B Engineered for autocomplete, streaming chat bubbles, classification, and summarization tasks.
attenval-code
High AccuracyQwen 2.5 Coder, DeepSeek Coder Trained on modern programming languages, AST refactoring, and test suite generation.
attenval-smart
Deep ThinkingDeepSeek R1, QwQ 32B Solves complex algorithmic logic, rigorous math, system design, and competitive coding problems.
Two Balances, Zero Ambiguity
We treat compute credits with strict financial discipline. Earned promotional credits and purchased prepaid capacity are distinctly tracked in your wallet.
Promotional credits earned through verified attention are always consumed prior to purchased credits.
Completing any qualifying developer spotlight extends your entire earned balance for another 30 days.
Credits purchased with INR persist indefinitely until consumed by your API keys.
Prepaid Capacity. No Hidden Subscriptions.
Buy what you need when you're ready. All packs are one-time purchases with zero recurring charges and official GST invoices.
Ideal for hobbyists, prototyping new prompts, and running lightweight background scripts.
- ~4.5M input tokens (attenval-auto)
- No Attenval product expiry
- Full access to all 4 model aliases
- Instant GST receipt
For active developers building web apps, bots, and automated coding workflows.
- ~10.0M input tokens (attenval-auto)
- No Attenval product expiry
- Full access to all 4 model aliases
- High concurrency rate limits
- Instant GST receipt
Max value for heavy coding sessions, continuous agent loops, and deep reasoning tasks.
- ~27.0M input tokens (attenval-auto)
- No Attenval product expiry
- Full access to all 4 model aliases
- Priority routing failover
- Instant GST receipt
Calculate Your Monthly AI Credit Needs
Estimate token volume across model aliases and see exact credit and pack recommendations.
Zero Prompt Selling. Transient Processing by Default.
We do not log raw prompts or model completions to persistent database tables (`cf-aig-collect-log-payload: false`). Your data is processed in-memory solely to serve the API request. Statutory financial and GST audit records are preserved for 72 months in strict separation from AI telemetry.
Start building with verified AI compute today.
Sign up in 30 seconds, get your API key, and make your first completion with zero upfront payment required.