DeepSeek Reasoner (R1)
DeepSeekDeepSeek's reasoning model. Comparable to OpenAI's o1 but at much lower cost.
DeepSeek Reasoner (R1) Pricing
| Token Type | Price per Million |
|---|---|
| Input tokens | $0.550 |
| Output tokens | $2.19 |
Estimated Cost by Project Size
Realistic cost estimates for common coding scenarios. Assumes 30% cache hit rate where caching is available.
| Scenario | Token Usage | Estimated Cost |
|---|---|---|
| Small Script (1K lines) | 50K input / 30K output | $0.08 |
| Medium Feature (10K lines) | 500K input / 200K output | $0.63 |
| Large Project (50K lines) | 2,500K input / 1,000K output | $3.15 |
| Code Review (5K lines) | 250K input / 25K output | $0.15 |
Benchmark Performance — DeepSeek Reasoner (R1)
Third-party benchmark scores normalized to 0-100 scale. Higher is better. Aggregated scores from published third-party benchmarks. SWE-bench measures real GitHub issue resolution. LiveCodeBench measures competitive programming ability. HumanEval measures basic code generation. BigCodeBench measures practical, multi-step coding tasks. All scores normalized to 0-100 scale.
Sources: SWE-bench Verified, LiveCodeBench, HumanEval, BigCodeBench
Get Access to DeepSeek Reasoner (R1)
Ready to start using DeepSeek Reasoner (R1)? Get API access directly from DeepSeek.
How Does DeepSeek Reasoner (R1) Compare?
| Model | Input ($/M) | Medium Feature Cost | |
|---|---|---|---|
| DeepSeek Reasoner (R1) | $0.550 | $0.63 | selected |
| GPT-3.5 Turbo | $0.500 | $0.48 | Compare |
| Qwen 3 Coder | $0.500 | $0.57 | Compare |
| Qwen Plus | $0.400 | $0.38 | Compare |
| Mistral Medium | $0.400 | $0.54 | Compare |
| GPT-4.1 mini | $0.400 | $0.46 | Compare |
Related Models
Claude 3.5 Haiku
AnthropicFast, cost-effective model for high-volume tasks. Great for code review and simple queries.
Claude 3 Haiku
AnthropicCheapest Claude model. Fast responses for simple tasks and basic coding.
GPT-4o mini
OpenAIAffordable small model. Fast and cost-effective for high-volume coding tasks.
GPT-3.5 Turbo
OpenAIBudget model for simple tasks. Being phased out but still widely used.