base_model = "zhipuai/glm-4.6v"
reasoning_options = [{ type = "toggle" }]

[cost]
input = 0.3
output = 0.9
cache_read = 0.05

[limit]
context = 131_072