models --list
Models
Every model below is reachable through the same key and the same endpoint. Swap the `model` field -- nothing else changes.
$3/mo
Fast, cheap models for everyday app traffic.
$4/mo ($1 first month)
Mid-tier models plus community perks.
$10/mo, 7-day trial
Flagship thinking and agentic models.
Standard
StandardFast, low-cost models for everyday app traffic. Included with the $3/mo Standard tier.
haiku-4.5-agentic
Standard200k context · ~$0.25/M in, $1.25/M out
Low-latency agentic model tuned for fast tool-calling loops.
gpt-5.4-mini
Standard128k context · ~$0.15/M in, $0.60/M out
Compact general-purpose model for everyday app traffic.
glm-5
Standard128k context · ~$0.20/M in, $0.80/M out
General-purpose multilingual model with strong cost/performance balance.
qwen-3-coder
Standard256k context · ~$0.30/M in, $1.20/M out
Code-specialized model tuned for completion and refactors.
phi-4-reasoning
Standard64k context · ~$0.10/M in, $0.40/M out
Small footprint model with lightweight step-by-step reasoning.
Donator
DonatorMid-tier models with more headroom, plus community perks. Included with Donator ($1 first month, $4/mo regular).
minimax-m2.1
Donator256k context · ~$0.50/M in, $2.00/M out
Balanced general-purpose model for production workloads.
deepseek-r2
Donator128k context · ~$0.55/M in, $2.20/M out
Reasoning-first model, strong on math and multi-step logic.
mistral-large-3
Donator128k context · ~$0.60/M in, $2.40/M out
Enterprise-grade general model with wide task coverage.
kimi-k2
Donator1M context · ~$0.45/M in, $1.80/M out
Long-context model built for large document and codebase input.
command-r-plus-2
Donator128k context · ~$0.40/M in, $1.60/M out
Retrieval-augmented generation specialist for grounded answers.
SuperMaxes
SuperMaxesFlagship thinking and agentic models for the hardest tasks. Included with SuperMaxes ($10/mo, 7-day free trial).
sonnet-4.5-thinking
SuperMaxes500k context · ~$3.00/M in, $15.00/M out
Flagship extended-reasoning model tuned for agentic coding.
opus-4.5
SuperMaxes500k context · ~$5.00/M in, $25.00/M out
Top-tier flagship model for the hardest reasoning tasks.
gpt-5.4
SuperMaxes256k context · ~$4.00/M in, $18.00/M out
Frontier general-purpose flagship model.
gemini-3-pro-thinking
SuperMaxes1M context · ~$3.50/M in, $14.00/M out
Flagship multimodal model with extended reasoning traces.
llama-4-405b
SuperMaxes128k context · ~$2.00/M in, $8.00/M out
Largest open-weight frontier model, self-hosted at scale.
Swap any model name below into this call:
curl https://api.using.ai/v1/chat/completions \
-H "Authorization: Bearer $USING_AI_KEY" \
-d '{
"model": "REPLACE_WITH_MODEL_NAME",
"messages": [{"role": "user", "content": "hi"}]
}'Pick a tier, grab a key, ship.
Every model above answers to the same request. Start on Standard or trial the SuperMaxes flagships free for 7 days.