issues
search
UbiquitousLearning
/
mllm
Fast Multimodal LLM on Mobile Devices
https://ubiquitouslearning.github.io/mllm_website
MIT License
381
stars
46
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
refactor:: remove Layer Class `Split`, replace it with `Tensor::split`
#136
yirongjie
closed
5 days ago
0
APk BUild issues
#135
Vinaysukhesh98
opened
1 week ago
0
Failed to allocate memory error on Galaxy S24 NPU
#134
gingerly
opened
1 week ago
0
运行时出现merge file is broken
#133
mailonghua
closed
1 week ago
2
feat: add MiniCPM 2B demo
#132
yirongjie
closed
1 week ago
0
关于OpPackage-LLaMAAdd中的Q6_V_valign_VVR计算方式的疑惑
#131
mailonghua
opened
1 week ago
0
refactor: `Tensor::run` &`Layer::getFunc`
#130
yirongjie
closed
1 week ago
0
fix: +-*/ for old front end
#129
yirongjie
closed
1 week ago
0
run “run_qwen_npu.sh “ fail. chip is Snapdragon 870
#128
yangh0597
closed
1 week ago
1
fix 修复windows环境
#127
WhiteNight123
closed
1 week ago
0
How did you obtain the two model files, qwen-1.5-1.8b-chat-int8.mllm and qwen-1.5-1.8b-chat-q4k.mllm?
#126
yhwang-hub
opened
2 weeks ago
3
./bin/main_qwen_npu fail
#125
zcxo
opened
2 weeks ago
2
./run_qwen-npu.sh failed
#124
zcxo
closed
2 weeks ago
2
./run_qwen_npu.sh失败
#123
zcxo
closed
2 weeks ago
0
perf: CPU Function: +-*/
#122
yirongjie
closed
2 weeks ago
0
refactor: `Tensor::run` &`Layer::getFunc`: Tensor& -> Tensor
#121
yirongjie
closed
3 weeks ago
0
refactor: `Layer::run` & `Tensor::getStaticFunc`
#120
yirongjie
closed
3 weeks ago
0
feat: add Phi-3-mini model
#119
WhiteNight123
closed
3 weeks ago
0
为什么预填充和解码不能都在 NPU 上运行？
#118
yhwang-hub
opened
3 weeks ago
4
Segmentation fault on OPPO FindX7 Ultra (Snapdragon8Gen3)
#117
bingo787
opened
3 weeks ago
1
Prefill speed is approximately 4~6 tokens/s for Qwen1.5-1.8B
#116
mengllm
opened
3 weeks ago
5
Crash on Xiaomi 14(8gen3) with QNN
#115
zhuipiaochen
opened
3 weeks ago
1
Android crashed and forcely rebooted when executing main_qwen_npu
#114
taegeonum
opened
3 weeks ago
9
doc: Update README.md
#113
xumengwei
closed
4 weeks ago
0
feat: Preliminary implementation on Qualcomm NPU (QNN) backend.
#112
liang1232018
closed
4 weeks ago
0
feat: llamafile_sgemm bias support
#111
chenghuaWang
closed
1 month ago
0
chore: Disable OpenMP for Mac.
#110
lx200916
closed
1 month ago
0
feat: GEMV + Bias mixed precision support for ARM Devices
#109
chenghuaWang
closed
1 month ago
0
feat: add clear_kvcache && fix: BUG in quantize.
#108
yirongjie
closed
1 month ago
0
fix: Qwen v1.5 Tokenizer bug
#107
chenghuaWang
closed
1 month ago
1
feat: add DEBUGSAVETENSOR & DEBUGOPTIME
#106
yirongjie
closed
1 month ago
0
feat: topk/topp sampling
#105
chenghuaWang
closed
1 month ago
1
perf: add AArch64 GEMM/GEMV for q4_0.
#104
yirongjie
closed
1 month ago
0
What is the difference between ggml and your project?
#103
Subuday
closed
1 month ago
4
Want to contribute. Looking for ideas.
#102
oddlyspaced
closed
1 month ago
2
perf: Use `vector<shared_ptr<Tensor>> Tensor::graphs`
#101
yirongjie
closed
1 month ago
0
feat: add Qwen 1.8B demo
#100
yirongjie
closed
1 month ago
0
feat:Add OPT support
#99
yirongjie
closed
1 month ago
0
feat: add elastic llama
#98
yirongjie
closed
1 month ago
0
Is Subgraph Heterogeneous Compute Available in MLLM?
#97
MaTwickenham
opened
1 month ago
2
[Question] Does this demo support mediatek neuropilot?
#96
AkaneTan
closed
1 month ago
2
doc: Update README.md
#95
yirongjie
closed
1 month ago
0
feat: Stablelm 2 1.6b support
#94
emt0re0
closed
1 month ago
0
Layer not found when trying mllm in Google Colab
#93
gsamaras
closed
1 month ago
9
feat: Add new demo: demo_imagebind_1mod
#92
yirongjie
closed
1 month ago
0
feat: Inference speed(tokens/s) profiling
#91
yirongjie
closed
1 month ago
0
feat: Add Multi-Head Latent Attention(MLA) support.
#90
yirongjie
closed
1 month ago
0
doc: Update README.md
#89
xumengwei
closed
1 month ago
0
feat: Yi-1.5-6B support
#88
chenghuaWang
closed
1 month ago
0
Update requirements.txt
#87
lx200916
closed
2 months ago
0
Next