GLM 5.3 Derisked
De-riskedFree to tryFewer refusals (18+): follows your instructions more closely. Vani's usage rules still apply.
GLM 5.3 Derisked is a Z.AI model with a 1M-token context window. It lets you choose how hard it thinks.
A de-risked model: fewer refusals for legitimate research, security, creative and technical work, for adults (18+). Our Terms of Service still apply. What de-risked means
Z.AI
3 models
GLM 5.3 Derisked in the Z.AI constellation, with GLM-5.3 and GLM 5.3 Flash.
A star's size and colour follow its API output price. Z.AI in the model list
Facts
- Context window
- 1M tokens
- Input
- Text
- Thinking effort
- Adjustable, up to High
- Speed
- Standard
- Available on
- All plans
- Intelligence rank at Vani
- #22
Based on data from Artificial Analysis (artificialanalysis.ai). The ranking is Vani's ordering of the models it offers and is not endorsed by Artificial Analysis. Source (opens in a new tab)
Guide
Best for
- Fiction for adults
- Frank answers
Good to know
- For adults (18+) only
- Much heavier on your allowance than standard models
GLM 5.3 Derisked API price
Use this model on the Vani API with the OpenAI- or Anthropic-compatible endpoints. Prepaid API prices per 1M tokens.
| Per 1M tokens | On Vani |
|---|---|
| Input | $3.90 |
| Output | $6.50 |
| Cached input | Same as input |
In Vani Chat it uses your plan's allowance: tokens per plan.
GLM 5.3 Derisked API quickstart
OpenAI-compatible: point any OpenAI SDK at https://api.vani.ai/v1 and send zhipu/glm-5.3-derisked as the model. The Anthropic-compatible /v1/messages endpoint works too.
curl https://api.vani.ai/v1/chat/completions \
-H "Authorization: Bearer $VANI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"zhipu/glm-5.3-derisked","messages":[{"role":"user","content":"Hello"}]}'import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.vani.ai/v1",
api_key=os.environ["VANI_API_KEY"], # a vani_ key
)
reply = client.chat.completions.create(
model="zhipu/glm-5.3-derisked",
messages=[{"role": "user", "content": "Hello"}],
)
print(reply.choices[0].message.content)import os
from anthropic import Anthropic
client = Anthropic(
base_url="https://api.vani.ai",
api_key=os.environ["VANI_API_KEY"], # a vani_ key
)
message = client.messages.create(
model="zhipu/glm-5.3-derisked",
max_tokens=1024,
messages=[{"role": "user", "content": "Hello"}],
)
print(message.content)Compare with
Z.AI
GLM-5.3
GLM-5.3 is a Z.AI model with a 1M-token context window. It reads images and lets you choose how hard it thinks.
1M context$3.08 output per 1M
Compare side by side with GLM-5.3OpenAI
GPT-6 Astra
GPT-6 Astra is an OpenAI model with a 1M-token context window. It reads images and lets you choose how hard it thinks.
1M context$35.00 output per 1M
Compare side by side with GPT-6 AstraAnthropic
Claude Fable 5
Claude Fable 5 is an Anthropic model with a 1M-token context window. It reads images and lets you choose how hard it thinks.
1M context$35.00 output per 1M
Compare side by side with Claude Fable 5
GLM 5.3 Derisked API: questions
- Is GLM 5.3 Derisked an uncensored model?
- People often look for it under that name. GLM 5.3 Derisked declines far fewer legitimate requests, because its trained refusal behaviour has been removed, but it is not a way around the law or our rules: the Terms of Service still apply, it is for adults (18+), and you are responsible for how you use what it generates.
- How do I use the GLM 5.3 Derisked API?
- Send zhipu/glm-5.3-derisked as the model to the OpenAI-compatible endpoint https://api.vani.ai/v1/chat/completions (or the Anthropic-compatible https://api.vani.ai/v1/messages), with a key from the Vani API Platform. Any OpenAI SDK works with base URL https://api.vani.ai/v1.
- How much does the GLM 5.3 Derisked API cost?
- On the prepaid Vani API, GLM 5.3 Derisked costs $3.90 per 1M input tokens and $6.50 per 1M output tokens. In Vani Chat it uses your plan's allowance.
Put a question to GLM 5.3 Derisked.
It is open in guest chat: no account needed.