issues
search
UbiquitousLearning
/
mllm
Fast Multimodal LLM on Mobile Devices
https://ubiquitouslearning.github.io/mllm_website
MIT License
537
stars
60
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Can VPTQ (an extreme-low bit quantization method) Be Adapted to MLLM?
#196
YangWang92
opened
1 day ago
0
feat: add Qwen1.5 1.8B Chat Android Demo.
#195
yirongjie
closed
1 day ago
0
fix: Android LFS.
#194
lx200916
closed
1 week ago
2
feat: support PhonLM-1.5B-Call demo in Android Demo
#193
yirongjie
closed
1 week ago
0
scripts: update build scripts
#192
yirongjie
closed
1 week ago
0
feat: QNN Multi Chunk Execution in New Frontend
#191
oreomaker
closed
1 week ago
0
Support PhoneLM decoding configuration
#190
liang1232018
closed
1 week ago
0
QNN support for other models (e.g., Gemma2, phi-2)
#189
Seoyoung-Ko
opened
1 week ago
0
feat: PhoneLM Instruct Android Demo.
#188
lx200916
closed
1 week ago
0
perf: i8 * i8 -> fp32 GEMM Boost
#187
chenghuaWang
closed
1 day ago
1
feat: add phi3v model
#186
k0zhevnikov
closed
2 days ago
0
fix: remove unused "fmt" files
#185
yirongjie
closed
1 week ago
0
1. refactor: add MLLM_LOG for Android support.
#184
lx200916
closed
1 week ago
0
Merge Branch "main" to "phonelm"
#183
yirongjie
closed
1 week ago
0
fix： NPU affect CPU
#182
yirongjie
closed
2 weeks ago
0
refactor: change tokenize method parameter from std::string& to const
#181
lx200916
closed
2 weeks ago
0
fix: smollm tokenizer regex pattern
#180
chenghuaWang
closed
2 weeks ago
0
feat: QNN New Frontend Phonelm Support and Refactors
#179
oreomaker
closed
2 weeks ago
0
how to upgrade qnn to 2.23.1.240531 to support v79
#178
yufeiy
opened
2 weeks ago
2
feat: drop xnn wrapper and move xnnwrapper to new front-end
#177
chenghuaWang
closed
2 weeks ago
1
fix: CPUTensorFunction.hpp
#176
UbiquitousLearning
closed
2 weeks ago
0
demo_qnn疑问
#175
ustcsq
opened
3 weeks ago
1
feat: Boost xnnpack backend inference speed by freeze tensor weight.
#174
chenghuaWang
closed
3 weeks ago
0
Add Bert for JNI.
#173
lx200916
closed
3 weeks ago
0
fix: BerTokenizer::tokenizes
#172
yirongjie
closed
3 weeks ago
0
fix: commen used in uni
#171
yirongjie
closed
3 weeks ago
0
feat:QNN New Frontend End to End Inference
#170
oreomaker
closed
3 weeks ago
0
fix: repe_theta is set wrong
#169
yirongjie
closed
3 weeks ago
0
Florence 2 & New Release APK
#168
Doocter
opened
3 weeks ago
2
refactor: migrate JNI to new backend.
#167
lx200916
closed
3 weeks ago
0
feat: Add modeling bert support
#166
XieWeikai
closed
3 weeks ago
0
feat: add PhoneLM
#165
yirongjie
closed
1 month ago
0
fix: mv `Tensor::graph` to `Module.activation_tensors`;
#164
yirongjie
closed
1 month ago
0
多模态推理比较慢
#163
dingtine
closed
3 weeks ago
3
refactor: add TransformerConfig
#162
yirongjie
closed
1 month ago
0
fix: CMakeLists.txt in `example`
#161
yirongjie
closed
1 month ago
0
feat: Support QWen2.5-1.5B, OpenELM-1,1B, DCLM-1B
#160
yirongjie
closed
1 month ago
0
Xnnpack backend support
#159
chenghuaWang
closed
3 weeks ago
0
QNN Module API(new frontend) Preliminary Support
#158
oreomaker
closed
4 weeks ago
1
fix: SmolLM name
#157
chenghuaWang
closed
1 month ago
0
Can mllm support MTK APU via NeuroPilot SDK?
#156
FranzKafkaYu
opened
1 month ago
2
Fixed typos.
#155
hustc12
closed
1 month ago
0
feat: add_profilling_activation
#154
chunfenri
closed
1 month ago
0
Error on Xiaomi 14(8gen3) with QNN
#153
ustcsq
closed
3 weeks ago
1
Xnnpack backend support
#152
chenghuaWang
closed
1 month ago
1
What is the difference between two llama-2 models?
#151
kjh-icnl
closed
1 month ago
2
Fix: QNN Cmakelists Config
#150
oreomaker
closed
1 month ago
0
fix: Memory Alloc bug in CPU Backend
#149
chenghuaWang
closed
2 months ago
0
feat: cross compile arm on windows(x86)
#148
chenghuaWang
closed
2 months ago
0
fix: windows compile bug
#147
chenghuaWang
closed
2 months ago
2
Next