issues
search
Infini-AI-Lab
/
Sequoia
scalable and robust tree-based speculative decoding algorithm
280
stars
29
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Work On CPU
#16
ZepinLi
opened
1 month ago
0
Question on tree search algorithm
#15
cyLi-Tiger
closed
2 months ago
3
Estimate the number of generated tokens per step from the acceptance-rate-vector?
#14
KexinFeng
opened
2 months ago
1
Reproducibility: the tree_search generates too small tree
#13
KexinFeng
opened
2 months ago
8
How to benchmark for speedup and acceptance rate?
#12
singularity-s0
opened
2 months ago
7
The support on vLLM?
#11
KexinFeng
opened
2 months ago
1
Is there any benchmark that compares Sequoia against vanilla speculative decoding?
#10
KexinFeng
closed
2 months ago
2
Thanks for your good work.
#9
xwjim
closed
3 months ago
0
paths fixed in tests/run_A100
#8
poedator
opened
3 months ago
0
Rotary fix
#7
poedator
opened
3 months ago
0
Fix datasets
#6
poedator
opened
3 months ago
0
Update README.md
#5
eltociear
closed
3 months ago
1
data loading timing and disk use
#4
poedator
opened
3 months ago
0
Integration with Lit-GPT
#3
tchaton
opened
3 months ago
2
Tensor shape mismatch when computing apply_rotary_pos_emb
#2
Tomorrowdawn
closed
4 months ago
5
Error `p.attn_bias_ptr is not correctly aligned` when testing
#1
poedator
closed
4 months ago
1