OpenBMB/MiniCPM

Issues

• [Feature Request]: 哭着求MiniCPM 4和Ollama的集成？

  #204 opened 16 days ago by fishfree
  15

• [Bug]: lora微调模型无法在vllm加载

  #241 opened 4 days ago by Caismis
  2

• [Bad Case]: 为什么推理速度比9b模型都要慢很多

  #191 opened 3 days ago by lixiaoyuan1029
  2

• 生成速度比Qwen2 7B还慢

  #202 opened 3 days ago by lucasjinreal
  6

• [Bad Case]: error

  #195 opened 4 days ago by lhjlhj11
  2

• [Bad Case]: ERROR occured when convert miniCPM3 model to GGUF format with llama.cpp, WHY?

  #226 opened 4 days ago by ridincal
  2

• [Bad Case]: MIniCPM3 原始pytorch .bin文件转为gguf失败

  #212 opened 14 days ago by sataliulan
  8

• [Feature Request]: Need Docker image for OpenBMB/llama.cpp

  #231 opened 4 days ago by bioinformatist
  1

• [Bug]: docker容器版本的sglang不支持minicpm3-4b

  #227 opened 4 days ago by cicicji
  1

• General Question: Which framework do you use to evaluate MiniCPM3-4B?

  #234 opened 4 days ago by xunkai55
  1

• 使用llamafactory微调报错

  #238 opened 5 days ago by lifelsl
  4

• [Feature Request]: 工具调用可以支持openai 的api function call吗

  #239 opened 6 days ago by lonngxiang
  5

• V1.0测试数据和脚本的链接失效

  #240 opened 5 days ago by MAxx8371
  0

• [Bug]: dataset.py文件conversation_to_ids方法input_ids调用shape属性错误

  #237 opened 8 days ago by KeepFaithMe
  1

• 请问embedding模型和rerank模型怎么finetune，用的FlagEmbedding的吗？

  #196 opened 12 days ago by LIUKAI0815
  2

• ImportError: cannot import name 'SamplerOutput' from 'vllm.sequence' (/root/miniconda3/lib/python3.11/site-packages/vllm/sequence.py

  #236 opened 8 days ago by badarrrr
  1

• [Bug]: WSL下安装 vllm報錯

  #232 opened 9 days ago by h122skite
  1

• Denis

  #235 opened 8 days ago by Denisdiabate22
  2

• [Bad Case]: 在机器上部署reranker模型之后，请求的时候报错了

  #230 opened 9 days ago by TOMATODA
  3

• [Feature Request]: MiniCPM-Embedding、MiniCPM-Reranker会基于MiniCPM3再发新版吗？

  #228 opened 9 days ago by dkasa
  1

• MiniCPM3微调

  #229 opened 9 days ago by lifelsl
  2

• MiniCPM可以微调做文本分类任务吗

  #219 opened 9 days ago by lifelsl
  1

• [Bad Case]: function calling可以不用vllm吗？

  #203 opened 16 days ago by cristianohello
  4

• [Bug]: openai_api_server_demo.generate_minicpm:158 - Input length larger than 20

  #211 opened 10 days ago by cxgreat2014
  2

• [Bad Case]: 运行Function call 简易实现代码出错，未调用相当的Function

  #218 opened 10 days ago by heiheiheibj
  1

• [Bad Case]: readme中llama.cpp的安装顺序是否应该调整

  #221 opened 10 days ago by SuperAZHE
  1

• [Bug]: 按照官方vllm装不上的解答步骤，未运行成功

  #222 opened 10 days ago by sssuperrrr
  3

• 能详细展开说一下提出的 “LLM x MapReduce” 么

  #225 opened 11 days ago by zhjunqin
  0

• 结构设计

  #223 opened 11 days ago by slZheng077
  0

• 持续预训练阶段的数据配比

  #220 opened 11 days ago by zyzyyy123
  0

• [Feature Request]: emebdding&reranker服务问题

  #207 opened 12 days ago by lyj157175
  3

• [Bug]: TypeError: 'ChatMessage' object is not subscriptable when MODEL_PATH=openbmb/MiniCPM3-4B

  #216 opened 12 days ago by cxgreat2014
  2

• [Bad Case]: python=3.11无法运行openai_api_server.py

  #215 opened 13 days ago by exthirteen
  5

• 可否提供一个 无限长上下文 MapReduce 的 sample code？

  #201 opened 16 days ago by msxfXF
  2

• What is llmxmapreduce? Any reference?

  #197 opened 16 days ago by world2vec
  1

• [Feature Request]: 请教关于显存和LLM x MapReduce的问题

  #208 opened 16 days ago by wciq1208
  3

• 看到这个logo我就直接加星了。

  #214 opened 13 days ago by WeileiZeng
  0

• [Feature Request]: 模型微调

  #213 opened 13 days ago by aph-asic
  0

• [Bug]: Model architectures ['MiniCPM3ForCausalLM'] are not supported for now.

  #199 opened 16 days ago by qq745639151
  4

• 请问如何配置上下文大于32k？

  #210 opened 16 days ago by thunder95
  0

• [Feature Request]: vllm 能直接用functioncall？

  #198 opened 16 days ago by lonngxiang
  8

• [Feature Request]: 工具调用微调

  #205 opened 16 days ago by douzi0248
  5

• [Bad Case]: function 多agent运行报错

  #206 opened 16 days ago by lonngxiang
  1

• [Feature Request]: 笔记本NPU如何调用MiniCPM

  #200 opened 16 days ago by R0k1e
  1

• [Feature Request]: 可否提供正確的CUDA version, PyTorch version, 以及 deepspeed version?

  #194 opened 21 days ago by joyyang1215
  1

• vllm运行你们给的demo需要多少显存

  #193 opened a month ago by lifelsl
  1

• 为了方便技术交流，拉了一个多模态大模型技术交流群，有需要的大家可以加入

  #192 opened a month ago by feihuamantian
  1

• 请教一下你们预训练用了1.1T tokens，花了多少GPU和时间

  #190 opened a month ago by zyh3826
  0

• [Bad Case]: 多模态 MiniCPM-V 推理报错

  #188 opened a month ago by c122-ode
  1

• 使用vllm推理时出现错误

  #189 opened a month ago by lifelsl
  2