unrefused-glm-5.3-flash
GLM-5.3 Flash, abliterated · mixture of experts · default in Chat
$3.00input
$0.30cached
$5.00output
Models
Open-weight, served under your policy.
Every model here is open-weight and licensed for commercial use. Pick by size and context; switch by changing one string. Prices are USD per million tokens and are placeholders until launch.
embed-base
Embed Base · embedding model
$0.02input
—cached
—output
Same request, any model
Model ids are stable strings. Change model and nothing else in your code changes: the request shape, tool calling, streaming and the policy decisions you get back are identical across the catalogue.
client.chat.completions.create(
model="unrefused-glm-5.3-flash", # the flagship — abliterated GLM-5.3
messages=msgs,
)Capacity and status
Each model runs as its own autoscaled deployment behind the gateway. Live availability per model is on the status page; GET /v1/models lists what your key can reach right now.