deepseek-ai/DeepSeek-R1
Parameters684B
Context160K
Input → outputtext → text
Licensemit
PrecisionF8_E4M3 · fp8
ArchitectureDeepseekV3ForCausalLM
ReleasedJan 20, 2025
Downloads1.3M
Other sizes and formats in the family
| Model | Params | Precision | Downloads |
|---|---|---|---|
| deepseek-ai/DeepSeek-R1-Zerofp8 | 685B | F8_E4M3 | 5.7K |
| deepseek-ai/DeepSeek-R1-Distill-Llama-70B | 70.6B | BF16 | 46.2K |
| deepseek-ai/DeepSeek-R1-Distill-Qwen-32B | 32.8B | BF16 | 368.4K |
| deepseek-ai/DeepSeek-R1-Distill-Qwen-14B | 14.8B | BF16 | 327.4K |
| deepseek-ai/DeepSeek-R1-Distill-Llama-8B | 8B | BF16 | 134K |
| deepseek-ai/DeepSeek-R1-Distill-Qwen-7B | 7.6B | BF16 | 385.1K |
| deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B | 1.8B | BF16 | 1.3M |
This release in the news
No coverage of this exact release yet.
Recent news about DeepSeek models
- Oct 19, 2026 · Open-source releasesgl-project/sglang v0.5.22
- Oct 18, 2026 · Opinion / analysisDeepSeek V4.1 Flash 1017 on 2x DGX Spark: 57 tok/s text, 92 tok/s code with TensorFold (2.1-2.3x the vLLM kit)
- Oct 17, 2026 · Tutorial / explainerjustxor/deepseek-local-guide-ru: 🐋 Полный гайд на русском: как установить и запустить DeepSeek локально (V4, R1) на ПК, Mac и сервере — Ollama, llama.cpp, LM Studio, vLLM, O
- Oct 16, 2026 · Research paperRethinking GRPO for Large Reasoning Model Safety Alignment
- Oct 14, 2026 · Open-source releaseNVIDIA/TensorRT-LLM v1.4.0rc0
- Oct 14, 2026 · Research paperBeyond the Trial-and-Error Loop: Hybrid Projection and Automated Tuning for Distributed Training
- Oct 13, 2026 · Research paperinLoRA-MINT: Internal Representations for Auditing Training Data in Domain-adapted LLMs
- Oct 11, 2026 · Opinion / analysisPSA: DeepSeek V4.1 Flash habitually exfiltrates API keys. It is dangerously misaligned and may be hazardous to use