wehub-resource-sync
22 lines
427 B
TOML
22 lines
427 B
TOML
# Opt in with chat_template_args.enable_thinking=true. Baseten documents no
|
|
# explicit false behavior, effort values, or reasoning-token budget.
|
|
# https://docs.baseten.co/inference/model-apis/reasoning
|
|
|
|
base_model = "zhipuai/glm-5.2"
|
|
name = "GLM 5.2"
|
|
|
|
[[reasoning_options]]
|
|
type = "toggle"
|
|
|
|
[interleaved]
|
|
field = "reasoning_content"
|
|
|
|
[cost]
|
|
input = 1.4
|
|
output = 4.4
|
|
cache_read = 0.3
|
|
|
|
[limit]
|
|
context = 202_720
|
|
output = 202_720
|