Issues
- 1
Question about the getdata() function
#59 opened by peterphancong - 1
- 1
About Up Sweep
#60 opened by haikunzhang95 - 2
Different grad of input between pscan and rnn
#61 opened by GlassyWing - 4
Bibtex
#55 opened by JulienSiems - 1
- 2
pscan speed compared to simple for loop
#57 opened by AnFreTh - 1
Question about converting mamba2 to onnx
#56 opened by ToaTao - 1
No module named 'mambapy.mamba_lm'
#54 opened by ooooma - 5
Values of deltaA are very large
#53 opened by anhtienng - 2
How to use cache in mamba2?
#52 opened by wwwqqyy - 2
flops about mamba2
#51 opened by dumpmemory - 4
MuP
#49 opened by norikazu99 - 1
Up sweep in parallel scan
#46 opened by anhtienng - 8
huge huge memory usage!!
#9 opened by eisneim - 2
Can I translate your PScan in Jax?
#45 opened by clementpoiret - 2
Default implementation of Jamba
#44 opened by erlebach - 1
Partial batches in Mamba_lm
#43 opened by erlebach - 2
Parallel pscan
#42 opened by erlebach - 4
support non-zero H[0] inputs
#38 opened by Zizzzzzzz - 3
Error in the Mamba Block forward function?
#41 opened by erlebach - 2
Question on using a sequence length > the max length I can hold in a batch due to memory usage for training
#40 opened by luke-mcdermott-mi - 5
Onnx export for the inference
#14 opened by llmexperiment - 5
Cuda Version
#37 opened by EddieEduardo - 1
Pscan documentation
#35 opened by poonam2308 - 3
- 3
- 4
[Feature Request] VideoMamba
#28 opened by MelihDarcanxyz - 5
delta question
#25 opened by AliYoussef97 - 3
Integration to `transformers`
#19 opened by ArthurZucker - 1
Working on AMD ROCm Platform
#22 opened by supersonictw - 0
- 4
A fresh can't start the model
#20 opened by Luchen-077 - 4
Possible SSM-Transformers implementation?
#18 opened by severian42 - 2
Why use element-wise multiplication rather than matrix multiplication in the function `selective_scan_seq`
#17 opened by cszhbo - 0
Segmentation fault with MLX
#16 opened by AshStill - 1
Support batch size > 1?
#15 opened by yjdy - 3
Mamba profiling_mamba.py script
#13 opened by llmexperiment - 3
- 9
About the speed test
#8 opened by MzeroMiko - 1
Can we get an explicit license?
#11 opened by CompRhys - 4
training functions?
#2 opened by win10ogod - 7
MLX inference error with BFloat16
#6 opened by beebopkim - 0
MLX memory usage at inference
#5 opened by alxndrTL