deepseek-ai/DeepSeek-V3-Base

Rename model-00016-of-000163.safetensors to 1a62629b4c871bf440ab008c1405c3e36d38458859c2a67fa4c7cd9a76a921fd

#79 opened 4 months ago by

Pq234

Delete README_WEIGHTS.md

#78 opened 4 months ago by

Pq234

Delete README_WEIGHTS.md

#77 opened 4 months ago by

Pq234

Delete README.md

#76 opened 4 months ago by

Pq234

Delete README.md

#75 opened 4 months ago by

Pq234

Fix chat_template crash when assistant message omits the `content` key

#74 opened 6 months ago by

qgallouedec

Deep

#73 opened over 1 year ago by

Whoami8888

Update inference/requirements.txt

#72 opened over 1 year ago by

Xiaocai123

ttest

#71 opened over 1 year ago by

Nellen

Make config params float to avoid warning in Transformers

#70 opened over 1 year ago by

Rocketknight1

Update README.md

1

#69 opened over 1 year ago by

Ratislav

【Q】shared_head weights of MTP

#68 opened over 1 year ago by

huang11

Update README.md

#67 opened over 1 year ago by

baishihao

Add library name and pipeline tag

#66 opened over 1 year ago by

nielsr

🚩 Report

#65 opened over 1 year ago by

didcnsrjf

DeepSeek LLM PEPE

🔥 1

#64 opened over 1 year ago by

Dethox

Request: DOI

#62 opened over 1 year ago by

ljh88301065

Update README.md

#61 opened almost 2 years ago by

chanhks55

`aux_loss_alpha` should be 1e-4 instead of 1e-3?

#60 opened almost 2 years ago by

cuichenx

skript

#59 opened almost 2 years ago by

technikgolem

有人跑成功了吗，用的什么配置跑的

6

#58 opened almost 2 years ago by

xl343

how to infer with mtp?

👍 3

#57 opened almost 2 years ago by

duanyu

Resource Requirements for Running DeepSeek v3 Locally

5

#56 opened almost 2 years ago by

wilfoderek

Update README.md

#55 opened almost 2 years ago by

malekradwan130

When running, the following error occurred, indicating that the config.json is missing.

#54 opened almost 2 years ago by

catsonthecar

Download link for models

2

#52 opened almost 2 years ago by

codecrypt112

viz

#51 opened almost 2 years ago by

vishnup13

Update README.md

#50 opened almost 2 years ago by

Aikun7777777

Update README.md

#49 opened almost 2 years ago by

Aikun7777777

Update modeling_deepseek.py

#47 opened almost 2 years ago by

erichartford

这东西cpu能跑不

8

#46 opened almost 2 years ago by

helloquark

Request: DOI

#45 opened almost 2 years ago by

fruit007

Delete .gitattributes

#44 opened almost 2 years ago by

Anna255

Delete .gitattributes

#43 opened almost 2 years ago by

Anna255

Create ？

#42 opened almost 2 years ago by

Anna255

Request: DOI

#41 opened almost 2 years ago by

abc123v9

Is there a way to convert it in to a 1 bit LLM? and use BITnet

➕ 2

3

#40 opened almost 2 years ago by

infinityai

Upload 01JG960S1ZEQEYRVHEYK9F55BT (1).jpg

1

#39 opened almost 2 years ago by

Rehanch3121

Hi!

👀 1

1

#37 opened almost 2 years ago by

Amirshahi2

Confusing Answer

👀 1

3

#36 opened almost 2 years ago by

Zilikon

You will release a small version for consumer hardware like the v2 generation?

9

#35 opened almost 2 years ago by

anon-linux-mint

deepseek v3 - the best open source AI model?

🔥 3

#33 opened almost 2 years ago by

Gerogiy

fp8转bf16的脚本在A100上无法执行

4

#32 opened almost 2 years ago by

duanyu

v3的自我认知怎么还不如之前

5

#31 opened almost 2 years ago by

yyren7

No monthly active user limitation on commercial use?

➕ 1

#30 opened almost 2 years ago by

DrNicefellow

Very impressive. Good world knowledge (SimpleQA of 25) despite high math/coding performance.

❤️ 3

2

#27 opened almost 2 years ago by

phil111

Converted bf16 Model on Hugging Face

❤️👍 5

#26 opened almost 2 years ago by

OpenSourceRonin

Create README.md

#24 opened almost 2 years ago by

xiaoshuai1234

应该把字节、阿里、百度的钱和显卡都分给deepseek，不然浪费资源啊

🤗 11

8

#23 opened almost 2 years ago by

eatcosmos

开源了也下不了。。哈哈哈

1

#21 opened almost 2 years ago by

hzxx0921

Rename model-00016-of-000163.safetensors to 1a62629b4c871bf440ab008c1405c3e36d38458859c2a67fa4c7cd9a76a921fd

Delete README_WEIGHTS.md

Delete README_WEIGHTS.md

Delete README.md

Delete README.md

Fix chat_template crash when assistant message omits the `content` key

Deep

Update inference/requirements.txt

ttest

Make config params float to avoid warning in Transformers

Update README.md

【Q】shared_head weights of MTP

Update README.md

Add library name and pipeline tag

🚩 Report

DeepSeek LLM PEPE

Request: DOI

Update README.md

`aux_loss_alpha` should be 1e-4 instead of 1e-3?

skript

有人跑成功了吗， 用的什么配置跑的

how to infer with mtp?

Resource Requirements for Running DeepSeek v3 Locally

Update README.md

When running, the following error occurred, indicating that the config.json is missing.

Download link for models

viz

Update README.md

Update README.md

Update modeling_deepseek.py

这东西cpu能跑不

Request: DOI

Delete .gitattributes

Delete .gitattributes

Create ？

Request: DOI

Is there a way to convert it in to a 1 bit LLM? and use BITnet

Upload 01JG960S1ZEQEYRVHEYK9F55BT (1).jpg

Hi!

Confusing Answer

You will release a small version for consumer hardware like the v2 generation?

deepseek v3 - the best open source AI model?

fp8转bf16的脚本在A100上无法执行

v3的自我认知怎么还不如之前

No monthly active user limitation on commercial use?

Very impressive. Good world knowledge (SimpleQA of 25) despite high math/coding performance.

Converted bf16 Model on Hugging Face

Create README.md

应该把字节、阿里、百度的钱和显卡都分给deepseek，不然浪费资源啊

开源了也下不了。。哈哈哈

有人跑成功了吗，用的什么配置跑的