LLM Inference API

OpenAI-compatible API powered by serverless GPUs. Pay per token, no minimums.

Start Building View Docs

Available Models

Llama 3.1 8B

Fast, efficient. Great for chat, summarization, and classification.

$0.10 / 1M input tokens $0.30 / 1M output tokens

Llama 3.1 70B

Powerful reasoning. Best for complex tasks and code generation.

$0.50 / 1M input tokens $1.50 / 1M output tokens

Mistral 7B

Compact and capable. Ideal for edge and high-throughput use cases.

$0.08 / 1M input tokens $0.24 / 1M output tokens

Kimi K3.1 Pro

Moonshot's flagship. Strong reasoning, long context, multilingual.

$0.60 / 1M input tokens $2.00 / 1M output tokens

Pricing

Pay only for what you use. No upfront costs, no minimum commitments.

Model Input (per 1M tokens) Output (per 1M tokens) Context
Llama 3.1 8B $0.10 $0.30 8K
Llama 3.1 70B $0.50 $1.50 8K
Mistral 7B $0.08 $0.24 32K
Kimi K3.1 Pro $0.60 $2.00 128K

Quick Start

curl https://api.logitbase.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama-3.1-8b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Drop-in replacement for OpenAI. Just change the base URL and API key.