litert-community/gemma-4-E2B-it-litert-lm

Any upcoming release for Gemma on Snapdragon Reality Elite quantized weights ?

#52 opened about 10 hours ago by

Kongkham

Running Gemma4 E2B on Mali GPU (RK3588)

#51 opened 9 days ago by

ProducerJenn

NPU support or compilation guidance gemma4 e2b on mediatek npu 780

#50 opened about 1 month ago by

Nicepotato01

Any plans to release a E4B G5 / G6 varient ?

👍 1

#49 opened about 1 month ago by

jegly

Please add the Qwen/Qwen3.5-0.8B model to your collection.

#48 opened about 2 months ago by

MT4030

Update E2B and E4B models

#47 opened about 2 months ago by

MT4030

Support for SM8735 NPU (Snapdragon 8s gen 4)

#46 opened about 2 months ago by

BarbocchioTheBearded

E2B language model update

#45 opened 2 months ago by

MT4030

Hello :)

#44 opened 2 months ago by

jegly

Does this base model support runtime LoRA sidecar? (LoraConfig loads but has no effect)

#42 opened 2 months ago by

eden-a

.safetensor to .litrtlm conversion

🚀 2

2

#40 opened 3 months ago by

sarvaraj12

Need Help Converting google/medgemma-1.5-4b-it

🔥 1

12

#39 opened 3 months ago by

larrybah

Add swift-litert-lm tag for iOS/Swift library integration

#38 opened 3 months ago by

mlboydaisuke

Create LOW_DEVICE_REVIEW.md

3

#36 opened 4 months ago by

ChoruYt

Using GPU backend results in precision loss

👍 1

2

#35 opened 4 months ago by

jadenyjw

NPU Inference Support for SM8845

4

#34 opened 4 months ago by

huajingzhou

NPU models with larger KV-cache (8K/16K/32K) for SM8845

1

#33 opened 4 months ago by

yellowredfish

Any chacne for a release of a 128k KV-CACHE version of the .litertlm gemma4 E2B-E4B models?

❤️ 2

2

#31 opened 4 months ago by

A-L-01

Model with QAT?

🤗 1

9

#30 opened 4 months ago by

4ntoine

How can I quantize/convert the model to support Qualcomm SM8850 platform?

👀 2

#29 opened 4 months ago by

jattanyiip

الاسفار المحذوفة

1

#27 opened 5 months ago by

FrBotros

does gemma-4-E2B-it_qualcomm_sm8750.litertlm run on NPU with chipset sm8750 , how to use it with android application?

2

#25 opened 5 months ago by

umakant-2026

Custom app with gemma model on android ,especially snapdragon gen 2

#22 opened 5 months ago by

vkvishal23

Model card

#21 opened 5 months ago by

QSCB

The latest model has no precision In GPU/CPU

#20 opened 5 months ago by

jeisonvendetta

gemma-4-E2B-it_qualcomm_qcs8275.litertlm vision encoder not optimized for NPU!

1

#19 opened 5 months ago by

sanoopsadheerthan

New gemma-4-E2B-it.litertlm broken!

4

#16 opened 5 months ago by

lktinhtemp

Convert finetuned google/gemma-4-E2B-it model to liteRT-LM

➕ 14

#15 opened 6 months ago by

SalihHub

Conversion from fine tuned merged model to litertlm

➕ 9

#14 opened 6 months ago by

darahask

Request for conversion flow of Gemma-4 Qualcomm NPU model

👀 2

4

#13 opened 6 months ago by

sanoopsadheerthan

does this litertlm support vision also, and how can i use for image on android application?

1

#11 opened 6 months ago by

piyushsinha0908

Upload 3 files

#10 opened 6 months ago by

sysia48

Are there plans to support Qualcomm 8550 NPU?

🤝 3

#8 opened 6 months ago by

MrrrKing

How to convert Gemma-4 model to litertlm format?

5

#7 opened 6 months ago by

hpkim0512

SM-8650 Model

1

#6 opened 6 months ago by

HarshTri007

How did you LiteRT Gemma4 ?

3

#5 opened 6 months ago by

Handyfff

GPU precision bug

2

#4 opened 6 months ago by

jeisonvendetta

litertlm for qcs6490

1

#3 opened 6 months ago by

suneetk