Catalog
Z.airouted

GLM-5.3-Flash Uncensored (NVFP4)

GLM-5.3-Flash Uncensored is a community build of GLM-5.3-Flash (dealignai, MIT licence) with refusal behaviour removed by a permanent weight edit — no fine-tuning and no prompt tricks; the vision input is kept. It was released for alignment and safety research, and its card reports no refusals on a standard harmful-request benchmark, so anything user-facing needs your own moderation. The model is currently offline; the card stays for reference.

model id: glm53-flash-unc

from openai import OpenAI client = OpenAI(    api_key="hb-...",    base_url="https://api.heabsy.com/v1",) r = client.chat.completions.create(    model="glm53-flash-unc",    messages=[{"role": "user", "content": "Hello"}],)print(r.choices[0].message.content)
Context
1M tokens
Serving
Global providers · EU contract
Tool calling
Reasoning
Structured output
Vision
Input
on request/ 1M
Output
on request/ 1M

Price list, updated 12 September 2026. No minimum spend, no subscription.

About the model

Developer
dealignai
Released
August 2026
Licence
MIT
Architecture
mixture of experts · 320B parameters, 18B active per token
Based on
GLM-5.3-Flash
Input
text, images
Reasoning
switchable per request
Languages named by the developer
English

Source: the developer's model card, as of 17 September 2026.

Licence: what it means for your company

MIT. A permissive open-source licence: you may use, modify and sell products built on the model, including commercially; when you redistribute the weights, the licence and copyright notices go with them.

A summary of the licence, not legal advice.

Technical details

Context window
1M tokens
Max output
32,768 tokens
Throughput limit
1.3M tok/min
Tokenizer
Other
Streaming
yes
Open weights
dealignai/GLM-5.3-Flash-UNCENSORED-NVFP4
Supported parameters
max_tokens
1 … 32768
temperature
0 … 2
top_p
0 … 1
frequency_penalty
-2 … 2
presence_penalty
-2 … 2
stop
array
seed
integer
Bought for you, billed by us

This model is fulfilled through global providers under our European contract: one DPA with an EU company, one invoice for every model in the catalog, no US counterparty on your paperwork. Compute may run outside the EEA — if you need strict EEA-only processing, use our own-hardware model.

Questions about this model

Where is GLM-5.3-Flash Uncensored computed?

Inside the European Economic Area, on hardware we operate. The catalog reports the processing region per model, so you can answer a supervisory authority without asking us.

Does GLM-5.3-Flash Uncensored store prompts and completions?

No. This model runs with zero data retention: request content is not written to durable storage and nothing is used for training.

How large is the GLM-5.3-Flash Uncensored context window?

1M tokens, with up to 32,768 tokens of output.

What does GLM-5.3-Flash Uncensored cost?

This model is quoted on request — write to us and we will price it for your volume.

Is there a throughput limit on GLM-5.3-Flash Uncensored?

Yes, 1.3M tokens per minute. Higher limits are a question of volume, not of principle.

Which languages does GLM-5.3-Flash Uncensored support?

The developer names English. Languages not on that list may still work, but they are not claimed — test on your own material first.

Under which licence is GLM-5.3-Flash Uncensored released?

MIT. A permissive open-source licence: you may use, modify and sell products built on the model, including commercially; when you redistribute the weights, the licence and copyright notices go with them.

Can reasoning be switched off in GLM-5.3-Flash Uncensored?

Yes. Per the developer, thinking can be switched on or off for each request.

Compare with similar models

ModelDeveloperParametersContextLicenceInput / 1M
GLM-5.3-Flash Uncensored (NVFP4)dealignai320B / 18B1MMITon request
GLM-5.3-Flash (NVFP4)Z.ai320B / 18B1MMIT$0.04
GLM-4.7-FlashZ.ai30B / 3B195KMIT$0.09
Qwen3.8 27B (NVFP4)Qwen (Alibaba)27B256KApache-2.0$0.40

30 more models behind the same key and the same invoice.

Browse the catalog