GLM-5.3 uncensored providers
by Z.ai753B parametersUpdated
Two providers in this directory serve GLM-5.3 uncensored: Silk Compute at $2.00 in and $4.00 out per 1M tokens, and Abliteration at $3.00 and $5.00. Abliteration sells it as Abliterated Large v2.
InputsText input
Lowest in / out price$2.00 / $4.00per 1M
Context1.05M
Providers2 of 7
Providers
Sorted by input price, read November 18, 2026.
| Provider | Sold as | Input /M | Output /M | Context | Access |
|---|---|---|---|---|---|
| Silk Compute | GLM-5.3 | $2.00 | $4.00 | 1.05M | Verified email |
| Abliteration | Abliterated Large v2 | $3.00 | $5.00 | 1M | Open sign-up |
| Not served by | |||||
| Morph | Not listed | ||||
| Unfil.AI | Not listed | ||||
| Featherless | Not listed | ||||
| Awan LLM | Not listed | ||||
| ModelsLab | Not listed | ||||
Data policy and payment
| Provider | Data policy | Payment | API formats |
|---|---|---|---|
| Silk Compute | Zero data retention | Card, Crypto | Chat, Messages, Responses |
| Abliteration | Not stored by default | Card | Chat, Messages, Responses |
Setup
Silk Compute takes OpenAI SDK requests at its base URL; GLM-5.3 on Silk Compute lists the limits.
Python
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.silkcompute.com",
api_key=os.environ["SILK_API_KEY"],
)
reply = client.chat.completions.create(
model="glm-5.3-uncensored",
messages=[{"role": "user", "content": "Hello"}],
)