← Back to Explorer
Pending

DeepSeek-V3 (2024)

DeepSeek-V3 is a large language model released by DeepSeek, released under the MIT license, with a claimed knowledge cutoff of 2024-12-26. Its weights, configuration, and tokenizer files have been cryptographically hashed and are queued for on-chain attestation by Weight Registry.

Description

DeepSeek's flagship MoE base model preceding R1, 671B total parameters with a fraction active per token.

Metadata

Type
LLM
Issuer
DeepSeek
License
MIT
Size
641.30 GB
Knowledge cutoff
2024-12-26
Source
https://huggingface.co/deepseek-ai/DeepSeek-V3
Revision
e815299b0bcbac849fa540c768ef21845365c9eb

Hugging Face stats

1,111,843 downloads on Hugging Face · 4,266 likes · pipeline: text-generation · library: transformers

transformerssafetensorsdeepseek_v3text-generationconversationalcustom_codeeval-resultstext-generation-inferenceendpoints_compatiblefp8

Weight Hashes

Weights file
163 shards (Merkle root) · 641.30 GB
SHA-2568f5b7332b69152e6d049dbf4c128e1ac6011f40a797a3d44f24ab3759b9aef6d
SHA3-256ef363e177a80dd24f38d47a63acb40ce9bbe73c577864b1004047700d8c9d915
BLAKE3723587d214232f1c23135d5dd56133834ba7a8439292f60dc1cd753e24554c7e
Config file
config.json · 1.62 KB
SHA-256cbf0b95dc614de208a109bb5fd4e7eed11385e9c68411d2c17db5319443035d9
SHA3-256c3795b62604c344e68f506facc786816825c5aa34a0ac408dfb5a6ef173fd957
BLAKE3f5fe5a82351b9018c7809f1669c4a54d83640fb843b4e12fb675cff52492ffd4
Tokenizer file
tokenizer.json · 7.48 MB
SHA-256621ac2e32d0dba658404412318818aaa8ce8cda492e59830109d8da6b517fb41
SHA3-2563fde231bbd24db1d0d0c6b9cb33eb80bf3b52f7c7de319eb83b2bfa6f9d7a3b8
BLAKE381fcbc9ddcf24735eb45d31d5d38821e0820750b3ec376a4e00415ed69d37b1e

Verify it yourself

Download the file(s) from the source below, then hash them locally (sha256sum <file> on Linux/macOS, certutil -hashfile <file> SHA256 on Windows) and compare the output against the hash(es) shown above.