Skip to content
Pareto Inference

Point Claude Code at Pareto's OpenAI-compatible API to run GLM 5.3 Flash, DeepSeek V4.1 Flash and MiMo V2.6 Flash on Pareto GPUs. Pay per token from one prepaid balance.

Talk to usGet started

Setup

Get an API key in the dashboard, then add this to your coding agent.

Set PARETO_API_KEY, then start Claude Code with these variables. Use the base URL without /v1. Set all four model variables, or Claude Code requests Claude models and fails with a 401.

TerminalShell
export ANTHROPIC_BASE_URL=https://api.paretoinference.com
export ANTHROPIC_AUTH_TOKEN="$PARETO_API_KEY"
export ANTHROPIC_MODEL=z-ai/glm-5.3-flash
export ANTHROPIC_DEFAULT_HAIKU_MODEL=z-ai/glm-5.3-flash
export ANTHROPIC_DEFAULT_SONNET_MODEL=z-ai/glm-5.3-flash
export ANTHROPIC_DEFAULT_OPUS_MODEL=z-ai/glm-5.3-flash
claude
Get started now
Full Claude Code guide

Models

GLM 5.3 Flash$0.06 input, $0.20 output and $0.012 cached input per 1M tokensDeepSeek V4.1 FlashOff-peak: $0.105 input, $0.42 output and $0.0021 cached input per 1M tokensMiMo V2.6 Flash$0.126 input, $0.252 output and $0.00252 cached input per 1M tokens

Coding agents

OpenClawOpenClaw configurationHermesHermes setup commandOpenCodeOpenCode configurationCodexCodex configurationpipi configurationClineCline base URLGooseGoose configurationAiderAider commandsContinueContinue configurationOMPOMP configurationCrushCrush configurationQwen CodeQwen Code commandsKiloKilo base URLZCodeZCode reasoning parameter mapping