Skip to content
AI Atlas
ModelActiveOpen weights

Updated 13 min ago · first seen 11 Sept 2026

model_01M294X79SXKE96ECDA77935CX

Parameters
355.7B
T2 · 22 min ago
Released
1 Jul 2025
T2 · 17 min ago
License
mit
T2 · 17 min ago

Specification

Release date
1 Jul 2025

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

Openness
open-weights

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 22 min agomedium

License
mit

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

Architecture
DeepseekV3ForCausalLM

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

Parameters
355.7B

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 22 min agomedium

Base model
deepseek-ai/DeepSeek-R1-0528

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

File size
403.1 GB

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

Hugging Face repo
amd/DeepSeek-R1-0528-MXFP4

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 22 min agomedium

Downloads
12,506

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 min agomedium

Likes
2

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 min agomedium

Gated
No

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

Last modified
2026-02-26T05:11:25+00:00

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 22 min agomedium

Downloads all time
190,052

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 min agomedium

Model type
deepseek_v3

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

Tags
safetensors, deepseek_v3, custom_code, 8-bit, quark

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

Weights dtype
BF16, U8

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 17 min agomedium

Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →

Provenance

Attributed facts

17

Source tiers

T217

Freshest observation

13 min ago

Conflicts

None