DeepSeek R1 by DeepSeek demonstrates strong general intelligence, outstanding mathematical reasoning. View detailed benchmark data including scores across coding, math, reasoning, speed, and cost metrics.
Specifications last checked against the provider on . Benchmark scores carry their own source, linked beside each number.
General Benchmarks
Human preference ELO from blind head-to-head votes
LMSYS / HuggingFaceArtificial Analysis composite intelligence score across 10 sub-benchmarks
Artificial Analysis (via OpenRouter)Massive Multitask Language Understanding professional benchmark
HuggingFace LeaderboardCoding Benchmarks
Artificial Analysis composite coding score
Artificial Analysis (via OpenRouter)Math Benchmarks
Reasoning Benchmarks
Artificial Analysis composite score for multi-step tool-using tasks
Artificial Analysis (via OpenRouter)Speed Benchmarks
Cost Benchmarks
Context Benchmarks
Available from 19 providers
The same model costs different amounts depending on who serves it. You can bring your own key for 6 of these — connect it here.
| Provider | Input / 1M | Output / 1M | Cached in | Context | Your key |
|---|---|---|---|---|---|
| iFlow | $0 | $0 | — | 128K | — |
| NanoGPT | $0.4 | $1.7 | $0.2 | 128K | — |
| D.Run (China) | $0.55 | $2.2 | — | 131K | — |
| Alibaba (China) | $0.574 | $2.294 | — | 131K | — |
| Eden AI | $0.7 | $2.4 | — | 164K | — |
| Hugging Face | $0.7 | $2.5 | — | 64K | Supported |
| Kilo Gateway | $0.7 | $2.5 | — | 64K | — |
| NovitaAI | $0.7 | $2.5 | — | 64K | — |
| OpenRouter | $0.7 | $2.5 | — | 64K | Supported |
| Amazon Bedrock | $1.35 | $5.4 | — | 128K | Supported |
| Azure | $1.35 | $5.4 | — | 164K | Supported |
| Azure Cognitive Services | $1.35 | $5.4 | — | 164K | — |
| Merge Gateway | $1.35 | $5.4 | — | 164K | — |
| Vercel AI Gateway | $1.35 | $5.4 | — | 128K | Supported |
| IO.NET | $2 | $8.75 | $1 | 128K | — |
| Abacus | $3 | $7 | — | 128K | — |
| Together AI | $3 | $7 | — | 164K | Supported |
| Qiniu | — | — | — | 128K | — |
| Snowflake Cortex | — | — | — | 128K | — |
DeepSeek R1 — Benchmark Scores Overview
Scores normalized to percentage scale for visual comparison. ELO scores mapped to 0-100 range (1100-1500).
Compare DeepSeek R1 With
DeepSeek R1 — Frequently Asked Questions
How intelligent is DeepSeek R1?
DeepSeek R1 scores 1355 on the Chatbot Arena ELO rating, making it a high-performing AI model. This score is based on blind head-to-head human preference voting.
How much does DeepSeek R1 cost?
DeepSeek R1 costs $0.70 per 1M input tokens and $2.5 per 1M output tokens. This is mid-range pricing for its capability level.
How fast is DeepSeek R1?
DeepSeek R1 generates output at 35 tokens per second, which is slower, prioritizing quality over speed compared to other models. The time to first token is 950 ms.
How good is DeepSeek R1 at coding?
DeepSeek R1 achieves 49.2% on SWE-bench Verified, demonstrating moderate real-world software engineering capability. This benchmark tests the model's ability to resolve actual GitHub issues.
How good is DeepSeek R1 at math and reasoning?
DeepSeek R1 scores 97.3% on the MATH benchmark (competition-level mathematics). It also achieves 71.5% on GPQA Diamond, a graduate-level science reasoning benchmark.
What is the context window of DeepSeek R1?
DeepSeek R1 has a context window of 64K tokens. This determines how much text, conversation history, and code the model can process in a single request.
Who created DeepSeek R1?
DeepSeek R1 was created by DeepSeek. It is classified as a open source model in our catalogue.
Is DeepSeek R1 open source?
Yes, DeepSeek R1 is an open-source model. The model weights are publicly available for download and self-hosting.