DeepSeek Prompts Cheatsheet - R1 Reasoning & V3 Chat Writing Techniques

Essential DeepSeek prompt templates organized by model — R1 reasoning rules, V3 chat tricks, codegen, structured output. Copy a template, fill the variables, and ship.

AI Prompts·30 commands·Last updated 2026-08-21

R1 Reasoning Rules (deepseek-reasoner) 5

No system prompt — put all instructions in the user message
R1 ignores the system role; instructions must go in user
Do not use few-shot examples; let the model reason alone
R1 reasons better unaided; few-shot distracts
Set temperature to 0.6
R1 official recommendation, balances stability and diversity
Give the question directly; trust the chain-of-thought
No "think step by step" needed; R1 thinks internally
Take the final answer, ignore the reasoning trace
R1 emits a long trace; use the last conclusion

V3 Chat Rules (deepseek-chat) 5

Use a system prompt to lock role and behavior
V3 supports system for tone-setting
Few-shot works — give samples for complex formats
V3 mimics the example structure
Creative tasks temperature 0.7-1.0, precise 0.3-0.5
Tune by task type
top_p 0.95
V3 default nucleus value
Enable JSON mode / function calling for structured output
V3 supports native structuring

Code Generation 5

V3: implement {req} in {lang} with error handling, comments, unit tests
V3 is good for direct codegen
R1: reason about the optimal {algorithm} and explain complexity
R1 suits algorithm/hard-logic inference
Specify framework and version to avoid hallucinated APIs
Explicit constraints cut hallucinations
Ask for a runnable snippet + usage example
Easy to verify directly
Review generated code for safety before use
AI code needs human sign-off

Chinese & Long Text 5

Prompt in Chinese directly; no need to translate to English
DeepSeek is strong in Chinese
128K context: chunk long docs and cite sources
Chunking is more stable for very long text
Summarize/extract/translate with V3
V3 has good throughput and instruction-following
Use R1 for deep analysis with enough context
R1 suits in-depth reasoning
Ask the model to cite paragraph locations
Easier to verify long-text claims

Templates 5

You are {role}, goal {goal}, constraints {constraints}. Task: {input}
V3 role+goal+constraint triplet
Reason: {question}. Give a conclusion and show the derivation.
R1 reasoning-chain template
Compare A/B in a Markdown table, dimensions: {dimensions}
V3 comparison template
Output strictly per this JSON Schema, no explanation: {schema}
V3 structured JSON template
From the material above, list 3 pros/cons with evidence
V3 multi-angle analysis template

Pitfalls 5

R1: no few-shot / no system
Weakens reasoning quality
Precise tasks: avoid high temperature
0.3-0.5 is steadier
Watch truncation on long outputs; split if needed
Ask in batches for very long results
R1 reasoning trace is not the final answer
Use only the conclusion section
For sensitive/factual asks, have the model flag uncertainty
Reduce hallucination risk

Tips

  • The key DeepSeek difference: R1 (reasoning) and V3 (chat) have opposite prompt rules — R1 forbids system/few-shot, V3 uses system to lock behavior and few-shot works. Mismatching model and rule drops quality sharply.
  • Prefer V3 for codegen (runnable snippets) and R1 for algorithm/hard-logic inference; pick the model by task.
  • For Chinese prompts just use Chinese — DeepSeek is as strong in Chinese as in English, no translation needed.
  • For structured data use V3 JSON mode + function calling; more stable and controllable than asking R1 for JSON.

Official References

Commands are compiled from the official docs below. Click to verify the latest usage.

Maintained by LaoHand

Publicly updated on Aug 21, 2026, continuously proofread against official docs.

Contact Us

Wrong command or description? Send us corrections, business inquiries or product feedback by email.

Contact Us