InfiniteHiP: Extending Language Model Context Up to 3 Million Tokens on a Single GPU
Paper
•
2502.08910
•
Published
•
143
•
6
Meshwa Patel
Meshwa
AI & ML interests
None yet
Recent Activity
liked
a model
24 days ago
hexgrad/Kokoro-82M
commented on
a paper
27 days ago
InfiniteHiP: Extending Language Model Context Up to 3 Million Tokens on
a Single GPU
new activity
28 days ago
hexgrad/Kokoro-82M:v0.19 and v1.0 versions Comparison: Advancements and Limitations
Organizations
None yet
Meshwa's activity
commented
a
paper
27 days ago
v0.19 and v1.0 versions Comparison: Advancements and Limitations
4
#121 opened 30 days ago
by
MaverickBuffoon
UD-IQ1_M models for distilled R1 versions?
3
#6 opened about 1 month ago
by
SamPurkis
Curb Your Expectations: Data & Architecture Limitations
13
#69 opened about 2 months ago
by
hexgrad
[Closed] Jan 21 Update
52
#36 opened about 2 months ago
by
hexgrad
Add support for artefacts like laughs, whispering voice etc.
1
#62 opened about 2 months ago
by
Meshwa
How to create custom voice pack?
5
#53 opened about 2 months ago
by
Meshwa
How to fine-tune?
3
#15 opened 2 months ago
by
Meshwa
commented
a
paper
2 months ago
Librarian Bot: Add language metadata for dataset
#2 opened 2 months ago
by
librarian-bot
Is there a way to add new tokens to the tokenizer?
#14 opened 3 months ago
by
Meshwa
commented
2
papers
3 months ago
Running this on cpu requires flash_attn ! but we cant install flash_attn on cpu
18
#4 opened 9 months ago
by
Meshwa
Can I use SDXL LoRA with this model?
#4 opened 8 months ago
by
Meshwa
Love the project, when will CFG hyper-sd come?
12
#24 opened 11 months ago
by
brandostrong
Update app.py
#1 opened over 1 year ago
by
Meshwa