DeepSeek-R1 is a 671B parameter Mixture-of-Experts (MoE) model with 37B activated parameters per token, trained via large-scale reinforcement learning with a focus on reasoning capabilities. It incorporates two RL stages for discovering improved reasoning patterns and aligning with human preferences, along with two SFT stages for seeding reasoning and non-reasoning capabilities. The model achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks.
Web Site AI Model Web Page | |
Provider The entity that provides this model. | |
Chat Input a message to start chatting | - |
Release Date When the model was first released. | 1 year ago Jan 21, 2025 |
Modalities Types of data this model can process | text |
API Providers The providers that offer this model. (This is not an exhaustive list.) | DeepSeek, HuggingFace |
Knowledge Cut-off Date When the model's knowledge was last updated. | Unknown |
Open Source Whether the model's code is available for public use. | Yes |
Pricing Input Cost for processing tokens in your prompts | $0.55 per million tokens |
Pricing Output Cost for tokens generated by the model | $2.19 per million tokens |
MMLU Massive Multitask Language Understanding - Tests knowledge across 57 subjects including mathematics, history, law, and more | 90.8% Pass@1 Source |
MMLU-Pro A more robust MMLU benchmark with harder, reasoning-focused questions, a larger choice set, and reduced prompt sensitivity | 84% EM Source |
MMMU Massive Multitask Multimodal Understanding - Tests understanding across text, images, audio, and video | - |
HellaSwag A challenging sentence completion benchmark | - |
HumanEval Evaluates code generation and problem-solving capabilities | - |
MATH Tests mathematical problem-solving abilities across various difficulty levels | - |
GPQA Tests PhD-level knowledge in chemistry, biology, and physics through multiple choice questions that require deep domain expertise | 71.5% Pass@1 Source |
IFEval Tests model's ability to accurately follow explicit formatting instructions, generate appropriate outputs, and maintain consistent instruction adherence across different tasks | 83.3% Prompt Strict Source |
SimpleQA Assessing the accuracy of simple questions | - |
AIME 2024 | - |
AIME 2025 | - |
Aider Polyglot Multilingual programming benchmark. | - |
LiveCodeBench v5 Benchmark for real-time programming | - |
Global MMLU (Lite) A simplified version of the benchmark for assessing the universality of models at the global level. | - |
MathVista Evaluates the mathematical reasoning abilities of AI models within visual contexts | - |
Mobile Application | |
MathArena | |
| Avg. Score | 82% |
| AIME 2025 A test based on problems from the American Invitational Mathematics Examination, designed to assess the mathematical skills of models. | 89% |
| HMMT February 2025 A test based on problems from the Harvard-MIT Mathematics Tournament, February 2025, designed to assess the mathematical skills of models. | 77% |
| BRUMO 2025 | 92% |
| SMT 2025 A test based on problems from the Stanford Math Tournament, 2025, designed to assess the mathematical skills of models. | 83% |
| CMIMC 2025 A test based on problems from the Canadian Mathematical Olympiad, 2025, designed to assess the mathematical skills of models. | 69% |
Compare AI. Test. Benchmarks. Mobile Apps Chatbots, Sketch
Copyright © 2026 All Right Reserved.
Comments (1)
Jacquie
27 January 2026Digiturk paketler icinden size en uygun uyeligi kolayca secebilirsiniz. https://digiturkpaketler.com/kampanyalar