LFM3-VL-2B
#31252

Merged

model : add LiquidAI/LFM3-1.7B and LiquidAI/LFM3-VL-2B #31252

ngxson merged 8 commits into ggml-org:master from tdakhran:tarek/lfm3/upstream/1-text-vl

model : add LiquidAI/LFM3-1.7B and LiquidAI/LFM3-VL-2B

68ed85ac

tests : set the SWA rope base of the LFM3 test model

4db6a29c

llama : enable recurrent state rollback for LFM3

eae14494

Apply batched suggestions from code review

99c1c53e

gguf-py : align the lfm3vl mapping comments

ead02869

llama : fix CRLF line endings from the review suggestions

de0f1c29

tdakhran requested a review 7 days ago

tdakhran requested a review from

pwilkin 7 days ago

tdakhran requested a review from

ggerganov 7 days ago

tdakhran requested a review from

CISC 7 days ago

github-actions added model

github-actions added testing

github-actions added mtmd

github-actions added conversion

CISC added highlight

CISC approved these changes on 2026-10-27

tdakhran marked this pull request as draft 7 days ago

ngxson approved these changes on 2026-10-27

tests : add the LFM3 Metal fusion baseline

3bdba723

llama : keep the LFM3-VL depthwise conv kernels in F32 when quantizing

0e158670

tdakhran marked this pull request as ready for review 7 days ago

ngxson merged 19cce691 into master 7 days ago

tdakhran deleted the tarek/lfm3/upstream/1-text-vl branch 7 days ago

Reviewers

ngxson

CISC

pwilkin

ggerganov

Assignees

No one assigned

Labels

model testing mtmd conversion highlight

Milestone

No milestone

llama.cpp model : add LiquidAI/LFM3-1.7B and LiquidAI/LFM3-VL-2B #31252 Merged

model : add LiquidAI/LFM3-1.7B and LiquidAI/LFM3-VL-2B #31252

llama.cpp
model : add LiquidAI/LFM3-1.7B and LiquidAI/LFM3-VL-2B
#31252

Merged