google/gemma-3-27b-it · Discussions

#62 opened 6 days ago by

Waisullah

Example of a wrong answer from an Image analysis

#61 opened 6 days ago by

vince62s

Test using M1 Max (64G) and Word

#60 opened 6 days ago by

gptlocalhost

what does "it" in the model name represent?

#59 opened 11 days ago by

jdsouza

Update README.md

#58 opened 11 days ago by

denizzhansahin

Gemma3ForCausalLM has no `_prepare_4d_causal_attention_mask_with_cache_position` method defined in its base modeling class

#57 opened 12 days ago by

buckeye17-bah

Inference speed slow?

#56 opened 13 days ago by

banank1989

Internal Server Error for url using vllm

#55 opened 13 days ago by

panyupj

Could not find the transformer layer class SiglipMultiheadAttentionPoolingHead in the model.

#54 opened 13 days ago by

neilwu

Keeps repeating itself.

rdtfddgrffdgfdghfghdfujgdhgsf

#53 opened 15 days ago by

Why MLP so tiny but vision part of the model works quite well

#52 opened 18 days ago by

CCRss

Using google/gemma-3-4b-it as asssitant model for speculative decoding does not work

#51 opened 24 days ago by

sayambhu

Update README.md

#50 opened 25 days ago by

beita6969

Processor Decode does not remove <end_of_turn> token

#49 opened 26 days ago by

gka-arch

SigLIP or SigLIP2 encoder?

7

#48 opened 26 days ago by

orrzohar

Updated model can't stop generation

#47 opened 26 days ago by

JackBAI

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

#46 opened 27 days ago by

junnei

Regarding the issue of errors in inference when converting to float16.

#45 opened 27 days ago by

LBJ6666

Please release the weights of Gemini 1.5 Flash

#43 opened about 1 month ago by

ZeroWw

Template error: syntax error: Conversation roles must alternate user/assistant/user/assistant/

#42 opened about 1 month ago by

stcat

Blank String in System Prompt causes Error - chat_prompt needs some fixes!

#41 opened about 1 month ago by

calycekr

Specify maximum context length in config

#40 opened about 1 month ago by

saattrupdan

Tokens generated per second

3

#39 opened about 1 month ago by

rameshch

Image tokenization

#38 opened about 1 month ago by

Marbuel

TPU inference

#37 opened about 1 month ago by

Arpit-Bansal

How to use the tokenizer of gemma-3

#36 opened about 1 month ago by

rishav09

Is there a possiblity for Google to change the license type of Gemma3 or its next release?

#33 opened about 1 month ago by

harisnaeem

/v1/chat/completions endpoint not working

#32 opened about 1 month ago by

PremkumarChandak

please release AWQ version

#31 opened about 1 month ago by

classdemo

evals (PT vs IT)

#30 opened about 1 month ago by

erichartford

Provide an access token

#29 opened about 1 month ago by

Tirth09

Where are the QAT releases for Gemma 3?

#28 opened about 1 month ago by

Downtown-Case

Hio

#27 opened about 1 month ago by

Azeliox

CUDA error: CUBLAS_STATUS_NOT_SUPPORTED

#25 opened about 1 month ago by

surak

Function calling with Gemma 3

5

#24 opened about 2 months ago by

jasonisaac

can not run with vllm

6

#23 opened about 2 months ago by

tankpigg

Simple questions too hard?

8

#22 opened about 2 months ago by

urtuuuu

ImportError: cannot import name 'Gemma3ForConditionalGeneration' from 'transformers'

#21 opened about 2 months ago by

Sashv

awq version

8

3

#20 opened about 2 months ago by

rastegar

Adds `vocab_size` to config.json

#18 opened about 2 months ago by

clowman

ERROR：Gemma3Config' object has no attribute 'vocab_size'

15

#17 opened about 2 months ago by

nexyi

Languages list

#16 opened about 2 months ago by

averoo

problem with Quantization convertion. "SOLVED"

3

#15 opened about 2 months ago by

josef2600

Fail to Load Gemme3 27B

#14 opened about 2 months ago by

crm-ai

License

#13 opened about 2 months ago by

mrfakename

Quantization

#12 opened about 2 months ago by

Zelyanoth

Meet this error

7

5

#11 opened about 2 months ago by

crm-ai

Anybody thinking what I'm thinking?

#10 opened about 2 months ago by

darkc0de

Benchmarks on model card and report don't add up