Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / deepseek-ai/DeepSeek-V2 issues and pull requests
#92 - Inquiry about Key/Value Storage and Matrix Merging in DeepSeekerV2 Inference Code
Issue -
State: open - Opened by xlim1996 about 2 months ago
#91 - doc: followup #89 add client demo for using SGLang
Pull Request -
State: closed - Opened by Ying1123 about 2 months ago
- 1 comment
#90 - doc: followup #89 add client demo
Pull Request -
State: closed - Opened by zhyncs about 2 months ago
- 1 comment
#89 - doc: recommend SGLang for DeepSeek V2 inference
Pull Request -
State: closed - Opened by zhyncs about 2 months ago
- 1 comment
#88 - Function Calling比以前难触发了
Issue -
State: open - Opened by whoisfucker 2 months ago
- 5 comments
#87 - Exploring the Combined Effects of YaRN and Adjusted rope_base Values in deepseek v2
Issue -
State: open - Opened by hannlp 2 months ago
#86 - docs: fix incorrect link in README.md
Pull Request -
State: open - Opened by itaowei 2 months ago
#85 - Question about the design of bos and eos token
Issue -
State: open - Opened by jojo23333 3 months ago
#84 - 线上api如何稳定的触发 tool_calls
Issue -
State: open - Opened by wssnail 3 months ago
- 2 comments
#83 - ValueError: The model's max seq len (163840) is larger than the maximum number of tokens that can be stored in KV cache (13360). Try increasing gpu_memory_utilization or decreasing max_model_len when initializing the engine.
Issue -
State: open - Opened by ArtificialZeng 3 months ago
- 3 comments
#82 - empty response from server
Issue -
State: open - Opened by 879611427 3 months ago
- 2 comments
#81 - fix
Pull Request -
State: open - Opened by ArtificialZeng 3 months ago
#80 - HuggingFace中开源的代码似乎没有实现矩阵合并
Issue -
State: open - Opened by meteorlin 3 months ago
- 1 comment
#79 - 多轮在训练中是否需要特殊间隔符,用什么间隔符号?
Issue -
State: open - Opened by AceCHQ 3 months ago
#78 - DeepSeek-V2-Lite-Chat模型启动依赖问题
Issue -
State: open - Opened by Malowking 3 months ago
- 1 comment
#77 - 自配大模型服务器如何选择GPU,CPU和内存
Issue -
State: open - Opened by zhanghanting 4 months ago
#76 - Error executing method determine_num_available_blocks: vLLM multi node fails for both DeepSeek-Coder-V2-Instruct and DeepSeek-Coder-V2-Lite-Instruct
Issue -
State: open - Opened by liangfang 4 months ago
- 1 comment
#75 - 0628版本加载报错
Issue -
State: open - Opened by bestpredicts 4 months ago
#74 - 为何我在A800上运行DeepSeek-V2-Lite-Chat (SFT),竟然消耗60G的显存?!
Issue -
State: open - Opened by juhengzhe 4 months ago
- 3 comments
#73 - about the active param counts of DeepSeek-V2-Lite
Issue -
State: open - Opened by imhmhm 4 months ago
#72 - Why max_model_len only 8192 when inferencing with vLLM for DeepSeek-V2-Chat?
Issue -
State: open - Opened by ybdesire 4 months ago
#71 - 怎么用dspy里的方法来调用deepseek?
Issue -
State: open - Opened by buchikeke 4 months ago
- 2 comments
#70 - What's the Prompt and Response length in the Paper?
Issue -
State: open - Opened by JadeRay 4 months ago
#69 - Add support llama.cpp
Issue -
State: open - Opened by techn0man1ac 4 months ago
#68 - 希望做vs2022扩展
Issue -
State: closed - Opened by woaidianqian 4 months ago
- 1 comment
#67 - 怎么在langchain里面使用deepseek计算embedding?
Issue -
State: open - Opened by ShuoAndy 4 months ago
- 3 comments
#66 - 网页端的默认参数
Issue -
State: closed - Opened by JaheimLee 4 months ago
#65 - 您好,可以查看源码吗?
Issue -
State: open - Opened by Darleen71 5 months ago
#64 - 同一个请求连续多次尝试都是相同错误
Issue -
State: open - Opened by gauss-clb 5 months ago
#63 - docs: update README for LMDeploy support
Pull Request -
State: closed - Opened by zhyncs 5 months ago
- 1 comment
#62 - Main
Pull Request -
State: open - Opened by bang78945 5 months ago
#61 - 如何优化deepseek用来做文本审查时的prompt定义
Issue -
State: open - Opened by xfghvgnfyjssjgte 5 months ago
#60 - 如何让模型能够回答完问题自动停止
Issue -
State: open - Opened by hensiesp32 5 months ago
#59 - 关于DeepSeek-Coder-V2-Lite-Base的128k捞针测试结果
Issue -
State: open - Opened by chaochen99 5 months ago
- 1 comment
#58 - it swapped to chinese and i cant get it to change back to english
Issue -
State: open - Opened by james28909 5 months ago
- 1 comment
#57 - It won't answer questions about the events that transpired in Tiananmen Square from April 15, 1989, to June 4, 1989.
Issue -
State: open - Opened by richpav 5 months ago
- 4 comments
#56 - 128k的推理有例子吗?
Issue -
State: open - Opened by 520jefferson 5 months ago
- 2 comments
#55 - Will the Deepseek platform's API call be updated to support generating multiple texts (n>1)?
Issue -
State: open - Opened by zchuz 5 months ago
- 1 comment
#54 - Chat API响应的role字段不要设为null
Issue -
State: open - Opened by jichulu 5 months ago
- 1 comment
#53 - hi, could you provide a code like llama3?
Issue -
State: open - Opened by lambda7xx 5 months ago
- 2 comments
#52 - Compatibility issues with the OpenAI Python client.
Issue -
State: open - Opened by dennymao 6 months ago
- 2 comments
#51 - 敏感词封禁问题
Issue -
State: open - Opened by gauss-clb 6 months ago
- 2 comments
#50 - Knowledge cutoff date
Issue -
State: open - Opened by Shadow-Alex 6 months ago
#49 - 模型部署困惑
Issue -
State: open - Opened by ylhou 6 months ago
- 2 comments
#48 - Drop Token
Issue -
State: closed - Opened by Richie-yan 6 months ago
- 2 comments
#47 - 你好,现在不支持,计划支持函数工具调用吗?
Issue -
State: closed - Opened by cristianohello 6 months ago
- 1 comment
#46 - has it function calling?
Issue -
State: open - Opened by cristianohello 6 months ago
- 1 comment
#45 - has it function calling?
Issue -
State: closed - Opened by cristianohello 6 months ago
- 1 comment
#44 - docker for vllm. with deepseekv2 support merged
Issue -
State: open - Opened by supdizh 6 months ago
#43 - 有没有计划将 deepseek-v2-lite 上传到 modelscope
Issue -
State: closed - Opened by Tendo33 6 months ago
#42 - RuntimeError: mat1 and mat2 shapes cannot be multiplied
Issue -
State: open - Opened by tarrett 6 months ago
#41 - 缓存C<sup>KV</sup><sub>t</sub> 多卡并行推理是否需要每张卡缓存一份
Issue -
State: open - Opened by c-dafan 6 months ago
#40 - How to fine-tune deepseek v2 models?
Issue -
State: open - Opened by satheeshkatipomu 6 months ago
- 6 comments
#39 - 发送图片
Issue -
State: closed - Opened by 21JayChou 6 months ago
- 1 comment
#38 - 请增加gguf支持
Issue -
State: closed - Opened by jackbapa 6 months ago
- 1 comment
#37 - 服务器部署问题
Issue -
State: open - Opened by airsxue 6 months ago
- 2 comments
#36 - 太容易陷入死循环了
Issue -
State: open - Opened by rak-bn 6 months ago
- 1 comment
#35 - 如何能达到论文里说的吞吐量50000多tokens
Issue -
State: open - Opened by ly19970621 6 months ago
- 6 comments
#34 - Invalid max_token values
Issue -
State: open - Opened by audreyeternal 6 months ago
- 2 comments
#33 - 无法支持 autogpt 中的 langchain
Issue -
State: open - Opened by chenny 6 months ago
- 2 comments
#32 - 'detail': 'Content Exists Risk'
Issue -
State: open - Opened by 18534516725 6 months ago
- 3 comments
#31 - 偏好数据构造方法
Issue -
State: closed - Opened by pandaupc 6 months ago
- 1 comment
#30 - BadRequestError: Error code: 400 - {'detail': 'Content Exists Risk'}
Issue -
State: open - Opened by judeomg 6 months ago
- 6 comments
#29 - 当结尾 "finish_reason":"stop" 时,role 值为空
Issue -
State: open - Opened by yttchan 6 months ago
- 1 comment
#28 - Add MoE offloading strategy?
Issue -
State: open - Opened by Minami-su 6 months ago
#27 - How to understand W^UK can be absorbed into W^Q and W^UV can be absorbed into W^O?
Issue -
State: closed - Opened by cc752424640 6 months ago
- 1 comment
#26 - Comparison Between MLA and MHA in dense model
Issue -
State: open - Opened by mx8435 6 months ago
- 1 comment
#25 - Device-Level Balance Loss and Communication Balance Loss
Issue -
State: closed - Opened by hsm1997 6 months ago
- 1 comment
#24 - why i use vllm inference deepseek v2 ,speed is low
Issue -
State: open - Opened by ZzzybEric 6 months ago
- 2 comments
#23 - Failure to reproduce MLA > MHA
Issue -
State: open - Opened by faresobeid 6 months ago
- 5 comments
#22 - 代码开源相关
Issue -
State: closed - Opened by DXZDXZ 6 months ago
- 1 comment
#21 - Reproduce inference benchmark mentioned in the paper
Issue -
State: open - Opened by zhouheyun 6 months ago
- 4 comments
#20 - Error executing method determine_num_available_blocks
Issue -
State: open - Opened by empty2enrich 6 months ago
- 2 comments
#19 - MLA vs MHA
Issue -
State: open - Opened by jiangix-paper 6 months ago
- 1 comment
#18 - 如何在 langchain 中调用 DeepSeek-V2?
Issue -
State: closed - Opened by soloice 6 months ago
- 3 comments
#17 - docs: update README.md
Pull Request -
State: closed - Opened by eltociear 6 months ago
- 2 comments
#16 - Any plan to involve VQA
Issue -
State: closed - Opened by TheMattBin 6 months ago
- 1 comment
#15 - 量化
Issue -
State: closed - Opened by ccp123456789 6 months ago
- 1 comment
#14 - 请扩充模型的中文词表
Issue -
State: closed - Opened by sohowj 6 months ago
- 1 comment
#13 - About datasets
Issue -
State: closed - Opened by ftgreat 6 months ago
#12 - 如何实现Device limited route
Issue -
State: closed - Opened by dawson-chen 6 months ago
- 1 comment
#11 - 8 * A100 启动巨慢,有启动成功的勇士不
Issue -
State: closed - Opened by CarryChang 6 months ago
- 2 comments
#10 - Clarifications Needed on KVCache Compression and Matrix Operations in MLA KVCache
Issue -
State: open - Opened by hxer7963 6 months ago
- 1 comment
#9 - API ERROR
Issue -
State: closed - Opened by 851039536 6 months ago
- 4 comments
#7 - How to deploy in VLLM?
Issue -
State: open - Opened by ZHENG518 6 months ago
- 11 comments
#6 - Error in Equation 16?
Issue -
State: closed - Opened by zhongmz 6 months ago
- 1 comment
#5 - `V-MoE` token droping and `MoD`
Issue -
State: open - Opened by liyucheng09 6 months ago
- 8 comments
#4 - Could we have scores for `LongBookQA Eng` and `LongBookSum Eng`
Issue -
State: open - Opened by zxzzz0 6 months ago
#3 - Could we have an in4 model and its LiveCodeBench score?
Issue -
State: open - Opened by zxzzz0 6 months ago
- 1 comment
#2 - Can not use tool and function-call?
Issue -
State: open - Opened by edisonzf2020 6 months ago
- 26 comments
#1 - 请提供GGUF,并支持OLLAMA
Issue -
State: open - Opened by taozhiyuai 6 months ago
- 6 comments