models · deepseek-ai

DeepSeek-R1-Zero

DeepSeek-R1-Zero — RL-trained reasoning model without SFT warmup. Demonstrates emergent chain-of-thought.

What this registry has checked

Quarantine. We hold the publisher’s own published hash for these weights and have recorded it against this entry. We have not downloaded and hashed the weights ourselves, so the hash is the publisher’s claim, not our measurement.

Weight file SHA-256	bf4e7665d5edd7a80615864e108491b7a44cd9cdcb4134b5dac1d95f1ef8e63d
Method	hf-published-sha256

Re-check this yourself → Source → Full record →

What is recorded

Publisher	deepseek-ai
Origin	China
Licence	MIT
Parameters	684,531,386,000 (684.5B)
Architecture	deepseek_v3
Layers	61
Hidden size	7168
Attention	MLA (128 heads, 128 KV)
Native precision	F8_E4M3
Pulls recorded here	76,019
Downloads reported upstream	10,636

Reported scores

GPQA Diamond	73.3
AIME 2025	71

Reported by a third party and recorded here, not re-run by us.

Elsewhere in the registry

All models → Browse with filters → What will run on my hardware? →

Registry record as of 2026-06-11. Identity, licence and parameter count are properties of a release and do not move; tier can, and the verify link above re-checks it against the live registry rather than this page.