Resources

View closed (10)

Added Evaluation Benchmarks to Metadata

#34 opened 3 days ago by

mackenzietechdocs

which tool-call-parser should be used when serving by vllm?

#33 opened 3 days ago by

xiaoheixiaohei

DeepSeek V3.2 API returns only 0 and -9999 for logprobs

#32 opened 6 days ago by

xl00004

cool model

#31 opened 8 days ago by

lunarflu

Reproducibility inquiry

👀 🚀 3

#30 opened about 1 month ago by

andresnowak

DeepSeek v3.2

#29 opened about 1 month ago by

Diene10

如果基于sglang 部署如何支持思考/非思考模式切换？

#28 opened about 1 month ago by

verigle

Скарты

#27 opened about 2 months ago by

Mrdips

Upload Cadient Revenue Radar 2026.xlsx

#26 opened about 2 months ago by

basisakai

Question: Why are the definitions related to max-model-len in config.json and tokenizer_config.json inconsistent?

#25 opened about 2 months ago by

foyoux

Request: DOI

#24 opened about 2 months ago by

xtolxy1

Is it possible to run inference on an A100 GPU?

#23 opened about 2 months ago by

Tony664

3.2 Exp 32b or distilled Qwen ?

#22 opened about 2 months ago by

guizpublic

DeepSeek-V3.2 全方位最新实测出炉（300+维度），欢迎进群交流讨论~

#17 opened 2 months ago by

JEIN

Question about long-context evaluation in DeepSeek-V3.2-Exp

#15 opened 2 months ago by

fcMpKYz6Avp5QK

国庆deepwork

➕ 🤗 5

#14 opened 2 months ago by

fengyujian

能不能一直保留旧版的deepseek v3.1的API接口？

❤️ 👍 3

#10 opened 2 months ago by

lixin4sky

Full Coverage Video of V3.2 - Step by Step

👍 2

#9 opened 2 months ago by

fahdmirzac

The whale is back

❤️ 7

#8 opened 2 months ago by

Nechintosh

How Much VRAM ?

#7 opened 2 months ago by

Ni3SinghR

Transformers does not recognize this architecture

#6 opened 2 months ago by

eva20150932-atlascloud

Context length

#5 opened 2 months ago by

cheflee668

咱这个模型是非得国庆前更新吗？？

😔 👍 113

#1 opened 2 months ago by

luckjone