google/gemma-3-4b-it · Discussions

#76 opened 21 days ago by

tcy006

gemma 3

#75 opened about 1 month ago by

bojupexa

LoRA/PEFT can’t target nn.Parameter vision→text projection in Gemma-3 VLM — why use nn.Parameter instead of nn.Linear(bias=False)?

#74 opened about 2 months ago by

alexanderyj

4bit Quantization Failure for gemma 4b

#73 opened 2 months ago by

shabha7092

Gemma full

#72 opened 2 months ago by

RamarajuK

test

#71 opened 3 months ago by

munbujang

Inference Issue with Multi GPU's

#70 opened 3 months ago by

m-nameer

<pad> <pad> <pad> fix

#69 opened 3 months ago by

test53100

processor does not have a chat template

#68 opened 3 months ago by

Prabhjot410

onnx model

#67 opened 3 months ago by

Gerald001

Parameters are contradictory use_cache=False

#66 opened 3 months ago by

lemon0703

Re: failing structured output generation

#63 opened 4 months ago by

msi-sbraun-11

Cannot recreate benchmark results

➕ 1

3

#61 opened 4 months ago by

alternis

The example code provided (without modifications) produces gibberish

3

#60 opened 5 months ago by

zqQH

Update README.md

#57 opened 6 months ago by

dj1507

Missing values in 4B, 12B config.json

#56 opened 6 months ago by

depasquale

git-clone

#55 opened 6 months ago by

devrf

Update preprocessor_config.json

#54 opened 6 months ago by

nm-research

use model run failed

#53 opened 6 months ago by

pb-1

Bug report: torch._dynamo.exc.Unsupported: Unexpected type in sourceless builder transformers.models.gemma3.configuration_gemma3.Gemma3TextConfig

8

#51 opened 6 months ago by

YovelRVeritex

Update config.json

#50 opened 7 months ago by

olegshulyakov

Update README.md

#49 opened 7 months ago by

VANNVISAL

Error in loading Model

#46 opened 8 months ago by

sans07

gemma3-1b-it

#45 opened 8 months ago by

TAENNY

Explicitly include num_attention_heads and num_key_value_heads in text_config for Gemma3 4B

#44 opened 8 months ago by

ebsmothers

Deployment issues on sagemaker

#43 opened 8 months ago by

tychen6677

Please open-source Gemini 1.5 Flash

👍 2

#42 opened 8 months ago by

drguolai

CUDA error: device-side assert triggered

#41 opened 8 months ago by

arshiasoori

Low GPU Utilization during inference?

#39 opened 8 months ago by

BagelBig

Object detection capabilities ?

#38 opened 8 months ago by

syrineM

SigLIP or SigLIP2 encoder?

6

#37 opened 8 months ago by

orrzohar

text-generation

#36 opened 8 months ago by

rakmik

I'm releasing the speech version of Gemma-3!

❤️ 👍 6

8

#35 opened 8 months ago by

junnei

Update README.md

#34 opened 8 months ago by

sachin7yadava

Issue with vLLM Deployment of gemma-3-4b-it on Tesla T4 - No Output

#33 opened 8 months ago by

twodaix

Batch processing on a GPU?

#32 opened 8 months ago by

buckeye17-bah

Please release also Gemini Flash 1.5 weights.

#31 opened 9 months ago by

ZeroWw

VRAM not freed during long generations (Gemma, max_new_tokens=3000)

#29 opened 9 months ago by

Nessit

The checkpoint you are trying to load has model type `gemma3` but Transformers does not recognize this architecture

#28 opened 9 months ago by

bajajrahul03

Does not works with bitsandbytes 4bit and 8bit

#27 opened 9 months ago by

zokica

Pan&Scan is desable by default ?

3

#26 opened 9 months ago by

Jack-HR-Red

Upload tokenizer.model

#23 opened 9 months ago by

Nexesenex

Having a hard time fine tuning with LoRA

#21 opened 9 months ago by

abcampbell

Adding "vocab_size" to config.json

#20 opened 9 months ago by

nmysore

The checkpoint you are trying to load has model type `gemma3` but Transformers does not recognize this architecture.

5

#19 opened 9 months ago by

bedoonraj

getting key error while running inference for pipeline api given in "Usage section"