Rules

A shared tag with AI prompts and code snippets

From workspace: DeepSeek

Team: Main

Total snippets: 11

DeepSeek

Rules

11 snippets

Sampling parameters

Use temperature 0.5–0.7 and top-p 0.95 for stable output.

Set the temperature within the range of 0.5-0.7 (0.6 is recommended) to prevent endless repetitions or incoherent outputs. Also, a top-p of 0.95 is recommended.

Clear and specific prompts

Use plain language to clearly state what you want.

Write your instructions in plain language, clearly stating what you want. Complex, lengthy prompts often lead to less effective results.

Structure your prompt

Use tags or markdown to organize prompt sections.

Break up different parts of your prompt using clear markers like XML tags, markdown formatting, or labeled sections. This organization helps ensure the model correctly interprets and addresses each component of your request.

Forcing think

Use <think> if the model skips reasoning.

On rare occasions, DeepSeek-R1 tends to bypass the thinking pattern, which can adversely affect the model's performance. In this case, the response will not start with a <think> tag. If you see this problem, try telling the model to start with the...

Math tasks

Use ‘reason step-by-step’ + \boxed{} for math problems.

For mathematical problems, it is advisable to include a directive in your prompt such as: "Please reason step by step, and put your final answer within \boxed{}."

Clearly describe output

Define what makes the response correct or useful.

Paint a clear picture of your desired outcome. Describe the specific characteristics or qualities that would make the response exactly what you need, allowing the model to work toward meeting those criteria.

No system prompt

Avoid system messages, use only the user prompt.

Avoid adding a system prompt; all instructions should be contained within the user prompt.

Majority voting for responses

Generate multiple responses and pick the most common.

When evaluating model performance, it is recommended to generate multiple solutions and then use the most frequent results.

No few-shot prompting

Avoid examples, describe the task and output format instead.

Do not provide examples in the prompt, as this consistently degrades model performance. Rather, describe in detail the problem, task, and output format you want the model to accomplish. If you do want to provide examples, ensure that they align...

Set clear requirements

Explicitly state all task constraints and limits.

When your request has specific limitations or criteria, state them explicitly (like "Each line should take no more than 5 seconds to say..."). Whether it's budget constraints, time limits, or particular formats, clearly outline these parameters to...

No chain-of-thought prompting

Don't ask the model to 'reason step-by-step'—it does by default.

Since these models always reason prior to answering the question, it is not necessary to tell them to "Reason step by step..."

DeepSeek - Rules - AI Prompts & Code Snippets | Snippets AI