DeepSeek V4.1 Flash Abliterated vs GLM-5.3 Flash Uncensored
Prices per million tokens as of 2026-10-27. GLM-5.3 Flash Uncensored is the lower-priced build on a 2M-in, 500K-out month.
| DeepSeek V4.1 Flash Abliterated | GLM-5.3 Flash Uncensored | |
|---|---|---|
| Model id | deepseek-v4.1-flash-abliterated | glm-5.3-flash-uncensored |
| Kind | Abliterated | Uncensored |
| Developer | DeepSeek | Z.ai |
| Input per M | $0.8488 | $0.1086 |
| Cached input per M | $0.1455 | $0.0109 |
| Output per M | $1.6975 | $0.3802 |
| Context window | not published | not published |
| Max output | not published | not published |
| Last hour | Answering | Answering |
| Daily check | Steady since Oct 19 | Steady since Oct 12 |
| Worked month | $2.55 | $0.41 |
The month
- Input
- 2,000,000 tokens
- Output
- 500,000 tokens
- GLM-5.3 Flash Uncensored
- $0.41
- DeepSeek V4.1 Flash Abliterated
- $2.556.3× the other build
Same call, different id
Both answer the OpenAI wire format on one base URL and key. Nothing but the model id changes.
One request on each
curl https://api.uncensoredllm.ai/v1/chat/completions \ -H "Authorization: Bearer sk-ullm-live-…" \ -H "Content-Type: application/json" \ -d '{"model": "deepseek-v4.1-flash-abliterated", "messages": [{"role": "user", "content": "Hello"}]}'curl https://api.uncensoredllm.ai/v1/chat/completions \ -H "Authorization: Bearer sk-ullm-live-…" \ -H "Content-Type: application/json" \ -d '{"model": "glm-5.3-flash-uncensored", "messages": [{"role": "user", "content": "Hello"}]}'Questions
GLM-5.3 Flash Uncensored: $0.41 for a month of 2,000,000 input and 500,000 output tokens against $2.55 on DeepSeek V4.1 Flash Abliterated, 6.3× the price, at the catalog rates on 2026-10-27.
An abliterated build has the refusal direction removed from the weights; an uncensored build is a fine-tune or a system-level build that answers without declining. Both answer what the stock model refuses; the model pages name each build.
Yes. Both answer the OpenAI wire format on the same base URL and key; the model id is the only field that changes.
The daily real-model check: DeepSeek V4.1 Flash Abliterated reads "Steady since Oct 19", GLM-5.3 Flash Uncensored reads "Steady since Oct 12" at render.
Every pair is on the comparisons page; every model with all its prices is on the price index.