issues
search
InternLM
/
lmdeploy
LMDeploy is a toolkit for compressing, deploying, and serving LLMs.
https://lmdeploy.readthedocs.io/en/latest/
Apache License 2.0
3.08k
stars
277
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
How to quantify deepseek-ai/deepseek-vl-7b-chat
#1865
SunnyLee20230523
opened
6 hours ago
0
使用pipeline加载Qwen1.5-32B-Chat，tp=4，使用openai prompt格式提示其清洗中文但生成回复都是英文
#1864
Yang-bug-star
opened
8 hours ago
0
使用OpenAI format的输入得到的response要如何提取出回复文本，返回的response好像是分段的
#1863
Yang-bug-star
opened
8 hours ago
1
[Bug] 单轮的图文交错对话的实现原理
#1862
stay-leave
opened
9 hours ago
0
react test evaluation config
#1861
zhulinJulia24
closed
9 hours ago
0
Fix vl session-len
#1860
AllentDan
closed
5 hours ago
1
[side-effect] bring back "--cap" argument in chat cli
#1859
lvhan028
closed
7 hours ago
1
misc: update torch version range to 2.3.0
#1858
zhyncs
opened
12 hours ago
0
[Feature] update the range of torch versions
#1857
zhyncs
opened
12 hours ago
4
Support guided decoding for pytorch backend
#1856
AllentDan
opened
14 hours ago
0
feat: decouple input_ids and output_ids
#1855
zhyncs
opened
1 day ago
0
vision model use tp number of gpu
#1854
irexyc
opened
1 day ago
0
Optimize sampling on pytorch engine.
#1853
grimoire
opened
1 day ago
0
bump version to v0.5.0
#1852
lvhan028
opened
1 day ago
3
[Bug] python -m lmdeploy.serve.proxy.proxy --server_name "xxx" --server_port xxx --strategy "min_expected_latency"
#1851
zeroneway
closed
8 hours ago
1
misc: align PyTorch Engine temprature with TurboMind
#1850
zhyncs
closed
12 hours ago
1
[Bug] Segmentation fault: address not mapped to object at address 0x2058
#1849
austingg
opened
1 day ago
4
under stream mode, if break generator in advance, it may lead to server stuck [Bug]
#1848
shanekong
closed
8 hours ago
3
[Bug] InternLM2MLP.forward() missing 1 required positional argument: 'im_mask'
#1847
jiangjingz
opened
1 day ago
0
如何指定模型的数据类型为f16
#1846
Yang-bug-star
opened
1 day ago
1
Support phi3-vision
#1845
RunningLeon
opened
1 day ago
0
Maybe a workaround for qwen2 quantization Nan error
#1844
AllentDan
opened
1 day ago
3
[Bug] 多图推理效果不佳
#1843
stay-leave
closed
1 day ago
2
fix cogvlm vl template
#1842
RunningLeon
closed
1 day ago
3
[Bug] lmdeploy - [31mERROR[0m - Truncate max_new_tokens to 221
#1841
tairen99
opened
1 day ago
7
[Docs] 多模态模型的api_server应该如何多卡部署？
#1840
red-fox-yj
opened
2 days ago
5
[Feature] How to support bf16 when inferencing Internvl-chat
#1839
Leo-yang-1020
opened
2 days ago
1
Harden stream callback
#1838
lzhangzz
closed
2 days ago
0
fix image encoder request queue
#1837
irexyc
closed
2 days ago
0
[Bug] qwen2 awq量化微调后的模型报错
#1836
qiuxuezhe123
opened
2 days ago
8
使用TurboMind 推理 + Python 代码集成的方式报错
#1835
zhanghanweii
opened
2 days ago
5
[Bug] pytorch方式多卡部署internlm-xcomposer2-vl-7b，报错KeyError: 'parameter name can\'t contain "."'
#1834
zhaoning1987
closed
15 hours ago
4
[Feature] How to support do_sample config just like Automodel 能否像Automodel推理中的do_sample参数支持，支持使用确定性生成方法，而不是随机采样
#1833
Leo-yang-1020
opened
3 days ago
5
第一次合并测试
#1832
Nianqitongs
closed
3 days ago
0
[Bug] smoothquant量化Bacihuan2-7B-Chat模型，无法正常量化
#1831
CodexDive
opened
3 days ago
6
[Bug] Qwen-7B-Chat 量化报错 AttributeError: 'RMSNorm' object has no attribute 'variance_epsilon'
#1830
CodexDive
opened
3 days ago
1
Update engine.py to fix small typos
#1829
WANGSSSSSSS
closed
2 days ago
0
Nightly Build for LMDeploy
#1828
zhyncs
opened
4 days ago
4
Model name id returned is weird specially when using Docker [Bug]
#1827
Hugobox
opened
5 days ago
0
[Bug] awq for Qwen2-72B-instruct
#1826
Vincent131499
opened
5 days ago
17
compat internlm2 for pytorch engine
#1825
RunningLeon
closed
2 days ago
4
fix qwen-vl-chat hung
#1824
irexyc
closed
5 days ago
0
drop stop words
#1823
grimoire
opened
5 days ago
7
[Bug] smooth_quant量化后的模型重新运行，lmdeploy无法正常推理
#1822
CodexDive
closed
2 days ago
24
Fix Request completed log
#1821
irexyc
closed
5 days ago
3
Add Jetson platform support (by docker)
#1820
BestAnHongjun
opened
5 days ago
0
[Bug] MiniCPM-llama3-V2_5 启动后使用image url 使用base64 没有回复结果
#1819
weiminw
opened
5 days ago
3
[Bug] Task was destroyed but it is pending! ImageEncoder._forward_loop()
#1818
DefTruth
closed
4 days ago
5
[Feature] Option to also use host memory for the KV cache
#1817
josephrocca
opened
5 days ago
0
int8 kv cache 和 Flash Attention 无法一起使用
#1816
SeibertronSS
closed
6 days ago
2
Next