Instructions to use cyboghostginx/gemma-3-27b-it-Adetayo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cyboghostginx/gemma-3-27b-it-Adetayo with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="cyboghostginx/gemma-3-27b-it-Adetayo")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cyboghostginx/gemma-3-27b-it-Adetayo")
model = AutoModelForCausalLM.from_pretrained("cyboghostginx/gemma-3-27b-it-Adetayo", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cyboghostginx/gemma-3-27b-it-Adetayo with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cyboghostginx/gemma-3-27b-it-Adetayo"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cyboghostginx/gemma-3-27b-it-Adetayo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cyboghostginx/gemma-3-27b-it-Adetayo

SGLang

How to use cyboghostginx/gemma-3-27b-it-Adetayo with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "cyboghostginx/gemma-3-27b-it-Adetayo" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cyboghostginx/gemma-3-27b-it-Adetayo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "cyboghostginx/gemma-3-27b-it-Adetayo" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cyboghostginx/gemma-3-27b-it-Adetayo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cyboghostginx/gemma-3-27b-it-Adetayo with Docker Model Runner:
```
docker model run hf.co/cyboghostginx/gemma-3-27b-it-Adetayo
```

gemma-3-27b-it-Adetayo

Icelandic fine-tune of google/gemma-3-27b-it, tuned for Icelandic morphology and grammar.

Scores (Miðeind Icelandic LLM leaderboard, official run)

task	score
6-task average	68.23
Inflection (BÍN morphology)	96.54
5-task average (local, Miðeind lm-eval fork)	79.56
GED (grammatical error detection)	54.5
WikiQA-IS (knowledge)	11.58

Inflection is where the training went: supervised fine-tuning on recombined, decontaminated BÍN noun-phrase data, which generalises to reading and reasoning tasks it was never trained on. WikiQA is a capacity wall, not a tuning gap, and cheap continued pretraining does not move it.

Use

Standard text generation. Apply the Gemma chat template (tokenizer.apply_chat_template). Not a reasoning model, evaluate it answering directly.

License

Gemma derivative. Use is governed by the Gemma Terms of Use and the Gemma Prohibited Use Policy. "Gemma" is retained in the model name as required.

Training data and methodology are proprietary and are not distributed with the model.

Downloads last month: 4

Safetensors

Model size

27B params

Tensor type

BF16

Model tree for cyboghostginx/gemma-3-27b-it-Adetayo

Base model

google/gemma-3-27b-pt

Finetuned

google/gemma-3-27b-it

Finetuned

(447)

this model