Start
Models
Two uncensored models, one key. Both have their refusals removed from the weights, so they answer the first time without a jailbreak prompt.
Checked 26 October 2026
Side by side
Words and pictures
DeepSeek V4.1 Flash Uncensored
Reads pictures as well as text, answers at once, and thinks when you ask it to. For stories, games, research and agents.
- Model
deepseek-v4.1-flash-uncensored- Reads
- Text, pictures
- Context
- 1M tokens
- Longest reply
- 128K tokens
- Tools, thinking
- Yes · none, low, high, max
- Per 1M tokens
- $0.45 in · $1.75 out
For less
GLM 5.3 Flash Uncensored
Under half the price, for text. It thinks briefly before every answer.
- Model
glm-5.3-flash-uncensored- Reads
- Text
- Context
- 128K tokens
- Longest reply
- 32K tokens
- Tools, thinking
- Yes · low, high
- Per 1M tokens
- $0.20 in · $0.65 out
Which one to use
| If you need | Use |
|---|---|
| Pictures in the conversation | deepseek-v4.1-flash-uncensored: it reads up to ten per request. GLM reads text only. |
| The lowest price for text | glm-5.3-flash-uncensored: $0.20 in · $0.65 out per 1M tokens. |
| A fast reply with no thinking | deepseek-v4.1-flash-uncensored: it answers at once unless you ask it to think. GLM always thinks briefly first. |
| A very long context | deepseek-v4.1-flash-uncensored: a million tokens. |
| Samplers like temperature or min_p | deepseek-v4.1-flash-uncensored. GLM takes only tools and reasoning_effort. |
The fields each one takes
| Field | DeepSeek | GLM |
|---|---|---|
temperature | Yes | Dropped |
top_p | Yes | Dropped |
stop | Yes | Dropped |
seed | Yes | Dropped |
frequency_penalty | Yes | Dropped |
presence_penalty | Yes | Dropped |
repetition_penalty | Yes | Dropped |
min_p | Yes | Dropped |
top_k | Yes | Dropped |
tools | Yes | Yes |
tool_choice | Yes | Yes |
parallel_tool_calls | Yes | Dropped |
response_format | Yes | Dropped |
reasoning_effort | Yes | Yes |
A field a model can’t take is dropped and named in the x-nokeep-ignored header, never guessed at. Every model takes model, messages, stream, stream_options and max_tokens.
DeepSeek V4.1 Flash Uncensored
DeepSeek V4.1 Flash with its refusal direction removed. It reads text and pictures, answers at once unless you ask it to think (reasoning_effort low, high or max), calls tools, and holds a million tokens of context. $0.45 in · $1.75 out per 1M tokens. Its page.
GLM 5.3 Flash Uncensored
GLM 5.3 Flash with its refusal direction removed. Text only, under half DeepSeek’s price, with a brief think before every answer (reasoning_effort low by default, or high). $0.20 in · $0.65 out per 1M tokens. Its page.
GLM’s host can be busy at peak times. A busy call answers 503 busy with Retry-After: 5 and costs nothing.
Questions
Are these the same models as DeepSeek V4.1 Flash and GLM 5.3 Flash?
The same weights, with the refusal direction removed. They answer like the originals on everyday work and stop declining what they’re asked.
Will you add more models?
Only ones whose host keeps nothing and lets us serve them this way. New models appear in GET /v1/models and the changelog.