# Pricing and endpoint (Chat Completions): https://opencode.ai/v2/docs/console/models/ (accessed 2026-10-21).
# Mid-conversation flags left out: nvidia has no provider file for nvidia/nemotron-3-ultra-550b-a55b to follow (Zen passes requests through).
schema = 2
base_model = "nvidia/nemotron-3-ultra-550b-a55b"
name = "Nemotron 3 Ultra Free"
status = "deprecated"

[capabilities]
reasoning = { supported = true }

[api.chat-completions]
base_url = "https://opencode.ai/inference/openai/v1"
compatibility = { reasoning_field = "reasoning_content" }

[cost.text.input]
price = 0
cache = { read = 0 }

[cost.text.output]
price = 0