# Pricing and endpoint (Chat Completions): https://opencode.ai/v2/docs/console/models/ (accessed 2026-10-21).
# Effort: reasoning_effort = low|medium|high|xhigh|max (not re-verified on Zen: the free tier only serves OpenCode clients)
# Mid-conversation flags (system_updates, effort_updates, tools.updates) follow providers/reflection/models/beam-1.0-501b-a23b.toml (Zen passes requests through).
# Tools follow providers/reflection/models/Beam-1.0-501B-A23B.toml; Zen passes requests through.
schema = 2
base_model = "reflection/beam-1.0-501b-a23b"
name = "Beam Free"

[capabilities]
tools = { supported = true, strict = true, choice = ["auto", "none", "required", "tool"] }
reasoning = { supported = true, effort = ["low", "medium", "high", "xhigh", "max"] }
structured_output = 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