Frequently Asked Questions

AltLLM is a tiered, crypto-specialized AI infrastructure with a proprietary Knowledge Layer. It provides crypto-specific insights through MCP tools and offers access to multiple AI models through a single OpenAI-compatible API.

1. Sign up at the AltLLM Portal using Google OAuth. 2. Add credits to your account via Stripe. 3. Create an API key from the API Keys page. 4. Use the API key with our OpenAI-compatible endpoint at https://altllm-api.viber.autonome.fun/v1

AltLLM offers 27 models across 4 categories: General Purpose (altllm-mini, altllm-flash, altllm-light, altllm-standard, altllm-fast, altllm-pro, altllm-pro-max, altllm-max), Coding-Optimized (6 tiers with coding-optimized prompts), Native (altllm-native-* for direct provider access), and Hybrid Routing (semantic, tiered, cost, round-robin).

AltLLM uses prepaid credits with usage-based billing. You purchase credits in advance ($5, $10, $20, $50, or $100 packages), and usage is deducted based on the model tier you use. Prices are per 1M tokens with 100% margin over provider costs.

Mini: $0.20/$0.60 per 1M tokens (cheapest). Flash: $0.14/$0.80 per 1M tokens. Light: $0.50/$0.76 per 1M tokens. Standard: $0.60/$2.40 per 1M tokens. Fast: $0.80/$2.00 per 1M tokens. Pro: $1.20/$4.40 per 1M tokens. Pro-Max: $1.00/$6.00 per 1M tokens. Max: $2.00/$10.00 per 1M tokens (input/output).

No, credits do not expire. Your balance remains available until used.

Unused credits can be refunded within 24 hours of purchase. Contact support for refund requests.

Yes! You can enable auto-recharge in the Billing section to automatically top up your balance when it falls below a threshold.

Yes! AltLLM provides a fully OpenAI-compatible API. You can use the official OpenAI SDKs by changing the base_url to https://altllm-api.viber.autonome.fun/v1 and using your AltLLM API key.

POST /v1/chat/completions - Chat completions (streaming and non-streaming). GET /v1/models - List available models. GET /health - Health check endpoint.

MCP (Model Context Protocol) tools are automatically injected into every request. When you ask about crypto prices, tokenomics, or other crypto-related queries, the AI can use tools like get_token_price, get_project_alpha, or web_search without any configuration on your part.

Default rate limits are 60 requests per minute and 100,000 tokens per minute. Higher limits are available for enterprise customers.

Yes! Set stream: true in your request to receive Server-Sent Events (SSE) for real-time token streaming.

AltLLM includes 51 MCP tools: get_token_price (live prices via CoinGecko), get_project_alpha (investment thesis), get_tokenomics (token distribution), get_team_info (team profiles), web_search (real-time news), calculate_impermanent_loss (DeFi calculator), and get_gas_price (multi-chain gas).

The database includes 103 crypto projects across 8 categories: Layer 1, Layer 2, DeFi, Gaming, Infrastructure, Privacy, Meme, and Stablecoins.

Yes! Token prices are fetched from CoinGecko with a 60-second cache for performance. The AI will display the price along with 24-hour change percentage.

Go to the API Keys page in the portal, click 'Create new key', give it a name, and copy the key. The full key is only shown once, so save it securely.

Yes! You can create multiple API keys with different names to track usage across different applications or environments.

Contact support to request account deletion. All data including API keys, usage history, and billing information will be permanently removed.

Yes. API keys are hashed with bcrypt before storage. All communication uses HTTPS. We do not log or store your prompts or completions.

Hybrid routing intelligently selects the best model tier for each request. Instead of manually choosing a model, you can use altllm-hybrid-semantic and let the system route simple queries to cheaper models and complex queries to more capable models.

Semantic (routes by query complexity), Tiered (weighted random: 40% Light, 30% Standard, 20% Pro, 10% Max), Cost (always cheapest), and Turn (round-robin rotation).

Typical savings range from 25-50% depending on your query mix. Semantic routing provides ~30-40% savings, while Cost routing can save up to 50%.