DeepSeek-V3

deepseek-ai/DeepSeek-V3

Text Generation685B parameterstransformers
Tracked by OpenModelStats since September 6, 2026View on Hugging Face ↗
Downloads 30D
1.4M
Total downloads
21.8M
Gained 7D
+378.1K(increase)
1.8%(up) vs baseline
Gained 30D
—
Likes
4,401
+6 in 7D
Rank · tracked models
#295
3(up 3 places)this week
Parameters
685B
Last source update
Mar 27, 2025
Momentum
90.2
  • DeepSeek-V3 gained 378.1K downloads during the last seven tracked days.
  • It moved from #58 to #56 among tracked Text Generation models this week.

Downloads over time

Cumulative total downloads observed by OpenModelStats

Likes over time

Rank among tracked models

Lower is better

Overview

DeepSeek-V3 is a text generation model published by deepseek-ai. OpenModelStats has tracked the model since Sep 6, 2026, most recently observing it Oct 6, 2026.

Published
Dec 25, 2024
Last updated
Mar 27, 2025
Library
transformers
License
—
Task
Text Generation
Parameters
684,531,386,000
Spaces
100
Derivative models
1
Velocity 7D/day
54K
transformerssafetensorsdeepseek_v3text-generationconversationalcustom_codearxiv:2412.19437eval-resultstext-generation-inferenceendpoints_compatiblefp8region:us

Model family

Based on base-model relationships reported by the source's metadata.

Derivative models

Similar models

Models similar to DeepSeek-V3
ModelDL 30D
DeepSeek-V3-0324deepseek-ai1.4M
DeepSeek-V3.1deepseek-ai352.2K
DeepSeek-R1-0528deepseek-ai206K
DeepSeek-V3.1-Terminusdeepseek-ai22.4K
DeepSeek-V3.1-Basedeepseek-ai11.3K
DeepSeek-R1-Zerodeepseek-ai9,038