Tokedy — private LLM inference for businesses

Private LLM inference for businesses.One OpenAI-compatible API for leading language models. Your prompts and answers are never stored. Tokedy is an OpenAI-compatible inference API with a private, zero-retention engine, prompt caching, a choice of models, usage metering and per-token billing. Keep your OpenAI SDK and change the base URL and the API key. Models

DeepSeek V4 Pro (DeepSeek) · 1M context · model id deepseek-v4-pro

DeepSeek V4.1 Flash (DeepSeek) · 1M context · Text and images · model id deepseek-v4.1-flash

GLM-5.3-Flash (Z.ai) · 512K context · Text and images · model id glm-5.3-flash

PricingAPI usage pricing, billed per token. Per million tokens, in US dollars, as of October 23, 2026.

Model

Input

Cached input

Output

Context

DeepSeek V4 Pro

$0.528

$0.0176

$1.584

DeepSeek V4.1 Flash

$0.24

$0.024

$0.96

GLM-5.3-Flash

$0.12

$0.012

$0.40

512K

You pay from a prepaid balance for the tokens a model reads and writes. Cached input costs less, a request that fails on our side costs nothing, and there are no seats, minimums or contracts. Questions: support@tokedy.ai. Terms · Privacy · DPA