llama.cpp
llama : support LiquidAI LFM2 hybrid model family
#14620
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
11
Changes
View On
GitHub
llama : support LiquidAI LFM2 hybrid model family
#14620
CISC
merged 11 commits into
ggml-org:master
from
Liquid4All:lfm2-upstream
model : LiquidAI lfm2 350M/700M/1.2B dense text-only
52b2da6e
Fix cache
85c79868
Set is_recurrent from h_head_kv
09410920
Use layer_types instead of full_attn_idxs
2ddfa277
github-actions
added
Nvidia GPU
github-actions
added
python
github-actions
added
ggml
Vaibhavs10
requested a review
from
ngxson
238 days ago
Vaibhavs10
requested a review
from
ggerganov
238 days ago
make flake8 happy
b76c0582
CISC
commented on 2025-07-10
CISC
requested a review
from
compilade
238 days ago
Address PR feedback
a22c09ac
tdakhran
requested a review
from
CISC
238 days ago
CISC
approved these changes on 2025-07-10
Merge remote-tracking branch 'upstream/master' into tarek/lfm2-upstre…
5a53cf76
ggerganov
commented on 2025-07-11
ngxson
commented on 2025-07-11
CISC
added
hot
Merge remote-tracking branch 'upstream/master' into tarek/lfm2-upstre…
009c64a7
compilade
commented on 2025-07-11
Address PR feedback 2
3cadfcef
Merge remote-tracking branch 'upstream/master' into tarek/lfm2-upstre…
949e3e28
tdakhran
requested a review
from
ggerganov
237 days ago
tdakhran
requested a review
from
compilade
237 days ago
tdakhran
requested a review
from
ngxson
237 days ago
ggerganov
approved these changes on 2025-07-11
ngxson
approved these changes on 2025-07-11
compilade
approved these changes on 2025-07-11
Support Lfm2ForCausalLM architecture name as well
92793845
CISC
merged
f5e96b36
into master
237 days ago
tdakhran
deleted the lfm2-upstream branch
236 days ago
Login to write a write a comment.
Login via GitHub
Reviewers
compilade
ngxson
ggerganov
CISC
Assignees
No one assigned
Labels
Nvidia GPU
python
ggml
hot
Milestone
No milestone
Login to write a write a comment.
Login via GitHub