Models / Llama 4 Maverick
Llama 4 Maverick
allais/llama-4-maverick
Natively multimodal MoE with 1M-token context window.
visionMoEvisionlong-context
Parameters
400B (17B active)
Context window
1000K tokens
Serverless $/1M
$0.176 in · $0.704 out
reference $0.22 · $0.88
Dedicated
8× H100-80G
sourced live, undercuts $46.4/hr
Use via API
curl http://localhost:3000/api/v1/chat/completions \
-H "Authorization: Bearer $ALLAIS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "allais/llama-4-maverick", "messages": [{"role": "user", "content": "Hello"}]}'