# Sources (accessed 2026-09-28):
# https://www.alibabacloud.com/help/en/model-studio/qwen3-7-flash
# https://www.alibabacloud.com/help/en/model-studio/model-pricing (Singapore)
# https://www.alibabacloud.com/help/en/model-studio/context-cache
# Toggle: enable_thinking true|false
# Budget: thinking_budget (integer reasoning tokens, maximum 262144)
# Explicit cache: read 10%, creation 125% of the input price in each tier.
base_model = "alibaba/qwen3.7-flash"
reasoning_options = [{ type = "toggle" }, { type = "budget_tokens", max = 262_144 }]

[interleaved]
field = "reasoning_content"

[cost]
input = 0.03
output = 0.13
cache_read = 0.003
cache_write = 0.0375

[[cost.tiers]]
tier = { type = "context", size = 32_000 }
input = 0.10
output = 0.40
cache_read = 0.01
cache_write = 0.125

[[cost.tiers]]
tier = { type = "context", size = 256_000 }
input = 0.20
output = 0.80
cache_read = 0.02
cache_write = 0.25