wehub-resource-sync
21 lines
438 B
TOML
21 lines
438 B
TOML
# Baseten documents opt-in via chat_template_args.enable_thinking=true, but no
|
|
# explicit false behavior, effort values, or reasoning-token budget.
|
|
# https://docs.baseten.co/inference/model-apis/reasoning
|
|
|
|
base_model = "moonshotai/kimi-k2.7-code"
|
|
temperature = true
|
|
[[reasoning_options]]
|
|
type = "toggle"
|
|
|
|
[cost]
|
|
input = 0.95
|
|
output = 4
|
|
cache_read = 0.16
|
|
|
|
[limit]
|
|
context = 262_000
|
|
output = 262_000
|
|
|
|
[modalities]
|
|
input = ["text", "image"]
|