Skip to content
Weights Not Walls
Menu

deepseek-ai

DeepSeek-R1

Reviewed
Size
684B parameterscounted from the safetensors header on the Hub
Licence
unknown, MIT LicenseOur reading, check the licence: You can use, change and redistribute the model for any purpose, including commercially. Keep the copyright notice and the licence text with copies. You do not have to share your changes.
On the Hub since
20 Jan 2025the date the repo was created
Context length
160k tokensfrom the model's config
Kind
Base model (no base model declared on the Hub)DeepseekV3ForCausalLM
On Hugging Face
deepseek-ai/DeepSeek-R11.6M downloads in the last 30 days, 14k likes

Scores

14 independent results on 14 eval sets
GPQA diamondRe-run by Epoch AI
  • default setting71.7%±3.126 May 2025, log

Epoch's name: DeepSeek-R1. Best closed model on this set: gpt-6.2-astra_max, 96.8% (OpenAI). All 341 results

MATH level 5Re-run by Epoch AI
  • default setting93.1%±0.731 Jan 2025, log

Epoch's name: DeepSeek-R1. Best closed model on this set: gpt-5-2025-08-07_high, 98.1% (OpenAI). All 108 results

OTIS Mock AIME 2024-2025Re-run by Epoch AI
  • default setting53.3%±7.526 Feb 2025, log

Epoch's name: DeepSeek-R1. Best closed model on this set: claude-fable-5_high, 100.0% (Anthropic). All 318 results

Epoch's name: DeepSeek-R1. Best closed model on this set: gpt-5-2025-08-07_high, 88.0% (OpenAI). All 72 results

ARC-AGIReported

Epoch's name: DeepSeek-R1. Best closed model on this set: claude-fable-5_max, 98.5% (Anthropic). All 318 results

ARC-AGI-2Reported

Epoch's name: DeepSeek-R1. Best closed model on this set: gpt-6-astra_max, 95.0% (OpenAI). All 299 results

BalrogReported

Epoch's name: DeepSeek-R1. Best closed model on this set: gpt-6-astra_max, 68.3% (OpenAI). All 41 results

CritPtReported

Epoch's name: DeepSeek-R1. Best closed model on this set: gemini-4-carbon_high, 36.0% (Google DeepMind). All 271 results

Epoch's name: DeepSeek-R1. Best closed model on this set: gpt-5-2025-08-07_medium, 97.2% (OpenAI). All 61 results

Epoch's name: DeepSeek-R1. Best closed model on this set: gpt-5-2025-08-07_medium, 86.0% (OpenAI). All 49 results

LiveBenchReported

Epoch's name: DeepSeek-R1. Best closed model on this set: gemini-2.5-pro-exp-03-25, 82.3% (Google DeepMind). All 54 results

Epoch's name: DeepSeek-R1. Best closed model on this set: claude-mythos-preview-early, 85.2% (Anthropic). All 49 results

SimpleBenchReported

Epoch's name: DeepSeek-R1. Best closed model on this set: claude-fable-5-2_unknown, 94.9% (Anthropic). All 125 results

WeirdMLReported

Epoch's name: DeepSeek-R1. Best closed model on this set: gpt-6-astra_promax, 93.6% (OpenAI). All 173 results

Claimed by the maker Claimed

The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.

Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.

Independent scores from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Each row links its own source. What the badges mean.

Run it yourself

file sizes read from the Hub 19 Nov 2026

Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 0.54 GB, worked out from the model's config (deepseek-ai/DeepSeek-R1), at 16-bit precision. Choose a machine above to see a verdict for each format. Method

GGUF for llama.cpp, Ollama, LM Studio and others

unsloth/DeepSeek-R1-GGUF (converted by unsloth), 47k downloads. Sizes are the real files on the Hub, all parts added up.

FormatSizeMemory neededGet it
Q3_K_M297.3 GBfile size, 7 parts298.3 GB297.3 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/DeepSeek-R1-GGUF:Q3_K_M
Download
hf download unsloth/DeepSeek-R1-GGUF --include "DeepSeek-R1-Q3_K_M/DeepSeek-R1-Q3_K_M-*-of-00007.gguf" --local-dir DeepSeek-R1-GGUF
Q4_K_M376.7 GBfile size, 9 parts377.7 GB376.7 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/DeepSeek-R1-GGUF:Q4_K_M
Download
hf download unsloth/DeepSeek-R1-GGUF --include "DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-*-of-00009.gguf" --local-dir DeepSeek-R1-GGUF
Q5_K_M442.7 GBfile size, 10 parts443.8 GB442.7 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/DeepSeek-R1-GGUF:Q5_K_M
Download
hf download unsloth/DeepSeek-R1-GGUF --include "DeepSeek-R1-Q5_K_M/DeepSeek-R1-Q5_K_M-*-of-00010.gguf" --local-dir DeepSeek-R1-GGUF
Q6_K513.0 GBfile size, 12 parts514.0 GB513.0 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/DeepSeek-R1-GGUF:Q6_K
Download
hf download unsloth/DeepSeek-R1-GGUF --include "DeepSeek-R1-Q6_K/DeepSeek-R1-Q6_K-*-of-00012.gguf" --local-dir DeepSeek-R1-GGUF
Q8_0664.3 GBfile size, 15 parts665.3 GB664.3 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/DeepSeek-R1-GGUF:Q8_0
Download
hf download unsloth/DeepSeek-R1-GGUF --include "DeepSeek-R1-Q8_0/DeepSeek-R1.Q8_0-*-of-00015.gguf" --local-dir DeepSeek-R1-GGUF
BF161250.1 GBfile size, 30 parts1251.1 GB1250.1 + 0.54 + 0.50
Commands
llama.cpp
llama-server -hf unsloth/DeepSeek-R1-GGUF:BF16
Download
hf download unsloth/DeepSeek-R1-GGUF --include "DeepSeek-R1-BF16/DeepSeek-R1.BF16-*-of-00030.gguf" --local-dir DeepSeek-R1-GGUF

MLX for Apple silicon

FormatSizeMemory neededRepo
MLX 3bitmlx-community/DeepSeek-R1-3bit312.6 GBfile size313.7 GB312.6 + 0.54 + 0.50Download page
MLX 4bitmlx-community/DeepSeek-R1-4bit390.7 GBfile size391.8 GB390.7 + 0.54 + 0.50Download page

Original weights size from the safetensors header, for vLLM, SGLang, Transformers

FormatSizeMemory neededRepo
F8_E4M3original weights641.1 GBfile size642.2 GB641.1 + 0.54 + 0.50
hf download deepseek-ai/DeepSeek-R1 --local-dir DeepSeek-R1
Other conversions (14; sizes not read)

The Hub tags 70 quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.

Run it by API

Hugging Face Inference Providers, fetched 19 Nov 2026, 13:58 UTC

1 provider serves this model through the Hugging Face router. Prices are US dollars per million tokens. Precision is what the provider reports; a lower precision than the original weights can change answers. Live first, then cheapest output.

ProviderInOutContextPrecisionTokens/sFirst tokenUptime 24hToolsJSON
novita$0.70$2.5063knot given221.1 s?yesyes

Through the router, with a Hugging Face token (the router bills your Hugging Face account). Change the name after the colon to pick another provider.

curl https://router.huggingface.co/v1/chat/completions \ -H "Authorization: Bearer $HF_TOKEN" -H "Content-Type: application/json" \ -d '{"model": "deepseek-ai/DeepSeek-R1:novita", "messages": [{"role": "user", "content": "Hello"}]}'

Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.

Fine-tunes of this model

from our snapshot of the Hub

Repos that declare this model as their base, most downloaded first. Listing is not a recommendation; unreviewed ones have not been checked.

Planned for this page Not built yet

Try it in private chat
A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
Fine-tune it
Fine-tuning a few base models from this page and publishing the result here.
Creators and earnings
Pages for people who publish fine-tunes and adapters of this model, and what they earn.
Fund and bounties
Bounties for new open models, paid from a public fund. How the fund will work.
Contribute compute
How to give compute to open model work, with Akash.

Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.