DeepSeek-R1 671B
A 671 billion-parameterHow many numbers the model learned during training — the usual rough measure of its size. mixture-of-expertsA model split into parts where only a few run for each word, so it answers faster than its size suggests. model using ~37 billion parameters per token from DeepSeek. Open reasoning model plus distilled dense models.
- Size
- 671B37B active
- Memory to run
- ~169 GBsmallest, 8K ctx
- Context
- 160Ktokens
- License
- PermissiveMIT License
- Updated
- 18 Sept 2026
Which version to use
- Variant DeepSeek-R1 · DeepSeekTrained to think step by step before answering; slower, stronger on hard problems.
Where it runs
Best-fitting version on each reference system, at an 8K contextHow much text the model can consider at once, counted in tokens — roughly ¾ of a word each..
Runs well 0
None of the reference systems.
Slowly (CPU or offload) 1
- A100 80GB server~21 tok/s est.
Benchmarks
Each benchmarkA fixed set of questions every model is given, so their scores can be compared on the same task. is developer-reported; prompts and settings differ between labs. Bars are relative to the best open result.
Variants & downloads
Each variant is a separate set of weights. Expand one for its downloads: quantizationStoring each of the model’s numbers with fewer bits, so the file is smaller and needs less memory. trades a little quality for a much smaller file, and the memory column adds the working memoryExtra space the model needs while it answers. It grows with the length of the conversation, on top of the file itself. a conversation needs on top.
Variant DeepSeek-R1Permissive
- What it is
- Trained to think step by step before answering; slower, stronger on hard problems.
- Publisher
- DeepSeek
- License
- MIT License · commercial use allowed
- Released
- 20 Jan 2025
- huggingface
- deepseek-ai/DeepSeek-R1
Contributions are not open yet, so there is nothing here from members.
| Quantization | Format | Bits | Download | Memory @ 8K | Publisher |
|---|---|---|---|---|---|
| Download FP8fp8 | safetensors | 8.1 | 633 GiB | DeepSeekdeepseek-ai/DeepSeek-R1 | |
| Download Q3_K_Mk quant | gguf | 3.91 | 305 GiB | Unsloth AIunsloth/DeepSeek-R1-GGUF | |
| Download IQ2_XXSi quant | gguf | 2.06 | 161 GiB | Unsloth AIunsloth/DeepSeek-R1-GGUF |
Architecture details
- Parameters
- 671B
- Active / token
- 37B
- Architecture
- Mixture of experts
- Layers
- 61
- Attention heads
- 128
- KV heads
- 128
- Head dim
- 128
- KV cache @ 8K (fp16)
- 0.54 GiB
- Max context
- 163,840 tokens
Measured performance
Throughput on specific systems and runtimes, with the source of each measurement.
Community results
Reviews
Lineage
How this model’s variants relate to each other and to other models.
Sources & history
Sources
- Hugging Face Hub (live source, 1 record, 18 Sept 2026)
- Mutinai illustrative fixtures (illustrative fixture, 1 record, 13 Sept 2026)
External identifiers
- huggingface deepseek-ai/DeepSeek-R1
Field history
- license = "mit" — Hugging Face Hub (current)
Timeline
- DeepSeek-R1 and distilled models released under MIT