deepseek-ai/DeepSeek-Coder-V2

Issues

• Clarification on Code Implementation in DeepSeek vs Llama

  #53 opened 3 months ago by dog14230pp
  0

• 硬件的硬性要求

  #51 opened 3 months ago by walkchen9
  1

• ollama Model is configured wrong: Double BOS

  #50 opened 3 months ago by Snowman-25
  0

• Model always responds in Chinese when using standard way of vllm + huggingface model

  #49 opened 3 months ago by iamhappytoo
  0

• Error: An unknown error was encountered while running the model

  #48 opened 3 months ago by vishnunuk
  1

• Questions about exceeding Max_Token

  #47 opened 4 months ago by Hoyeon-Yu
  0

• details about the paper

  #46 opened 4 months ago by Brucewuzhang
  0

• dependency parsing code, deduplication script

  #38 opened 5 months ago by wentinghome
  1

• File name char prefix used in training

  #45 opened 4 months ago by valerybugakov
  0

• 请问有考虑发布量化版本吗？

  #33 opened 5 months ago by ybdesire
  1

• issue bug

  #43 opened 4 months ago by healerwhy
  0

• 建议

  #42 opened 5 months ago by Geek-Bob
  0

• Hi, I have a question about fine tuning.

  #41 opened 5 months ago by Yindy07
  0

• What is the FSDP value for `fsdp_transformer_layer_cls_to_wrap`?

  #40 opened 5 months ago by Metaspectral
  0

• DeepSeek-Coder-V2-Lite model GPU/RAM requirement

  #11 opened 6 months ago by HashedViking
  4

• How to evaluate on defects4j?

  #39 opened 5 months ago by WentaoTan
  0

• OutOfMemoryError: CUDA out of memory on RunPod

  #17 opened 6 months ago by loyal812
  1

• Inference freezes using the recommended VLLM approach

  #29 opened 5 months ago by dhaneshsabane
  1

• Loading DeepSeek-Coder-V2-Instruct using transformers fails due to OOM on 8 * A100 (80GB)

  #30 opened 5 months ago by dhaneshsabane
  5

• 请问怎么把deepseek和知识库结合生成私有化的AI Agent

  #37 opened 5 months ago by zhanghanting
  0

• When will the vllm PR be merged to the main branch?

  #19 opened 6 months ago by zuxin666
  8

• Is the API pricing for the Deepseek v2 coder 230B model?

  #2 opened 6 months ago by Hangsiin
  4

• prompt for evaluating Code Fixing in readme3.3/paper4.3

  #36 opened 5 months ago by renmengjie7
  0

• How to build a fine-tuning dataset for code completion？

  #35 opened 5 months ago by FWLamb
  0

• Any changes on repo level concatenation?

  #22 opened 6 months ago by IQ179
  13

• VLLM推理报错 模型：DeepSeek-V2-Chat-0628

  #34 opened 5 months ago by zhangfan-algo
  0

• reinforcement learning improvement

  #31 opened 5 months ago by Ski-ing
  2

• 关于论文结果复现

  #32 opened 5 months ago by yingweima2022
  1

• FIM not working

  #27 opened 6 months ago by gergap
  3

• deepseek-coder-v2在使用到一定次数后开始疯狂胡乱输出。

  #23 opened 6 months ago by DoiiarX
  3

• Weird responses

  #28 opened 6 months ago by EduardDoronin
  0

• Requirements for the DS-Coder-V2-Instruct

  #26 opened 6 months ago by Urvesh71
  3

• Why DeepSeek-Coder-v2 236B is not trained with FIM objective?

  #18 opened 6 months ago by wasiahmad
  3

• 《开源大模型食用指南》sel-llm更新了DeepSeek-Coder-V2-Lite-Instruct模型的部署与微调教程！

  #21 opened 6 months ago by KMnO4-zx
  1

• How many tokens are generally trained in total?

  #10 opened 6 months ago by lengyueyang
  1

• 可以说明一下2024高考数学的评测方法和评测数据吗？

  #8 opened 6 months ago by Chenzongchao
  2

• CanNot Finetune deepseek-coder-v2-lite via modeling_deepseek.py

  #6 opened 6 months ago by chencyudel
  8

• Any plan to release the fintune example?

  #5 opened 6 months ago by SupercarryNg
  3

• mismatch between example code and model files

  #25 opened 6 months ago by yucc-leon
  1

• How is the result on SWE-bench obtained?

  #3 opened 6 months ago by zkx06111
  1

• What is the special token "<｜completion｜>" used for?

  #24 opened 6 months ago by yucc-leon
  0

• paper number not aligned

  #20 opened 6 months ago by lwaekfjlk
  0

• Any plans to release the 1B model as well?

  #4 opened 6 months ago by zuxin666
  1

• Model always responds in Chinese, ignores system prompts stating to only reply in English

  #12 opened 6 months ago by sammcj
  17

• Document tool calling capabilities

  #14 opened 6 months ago by goodov
  1

• Reward model in the reinforcement learning process

  #16 opened 6 months ago by bin123apple
  0

• about basemodel choice

  #15 opened 6 months ago by han508
  0

• continue.dev autocomplete integration

  #13 opened 6 months ago by symph-antonio
  0

• Model just solved RAS accounting problem correctly

  #9 opened 6 months ago by Priestru
  0

• exponential normalization technique

  #7 opened 6 months ago by futuristx
  0