Instructions to use trashpanda-org/gemma-4-31b-larkspur-v0.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use trashpanda-org/gemma-4-31b-larkspur-v0.5 with Transformers:

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

pipe = pipeline("image-text-to-text", model="trashpanda-org/gemma-4-31b-larkspur-v0.5")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("trashpanda-org/gemma-4-31b-larkspur-v0.5")
model = AutoModelForImageTextToText.from_pretrained("trashpanda-org/gemma-4-31b-larkspur-v0.5")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use trashpanda-org/gemma-4-31b-larkspur-v0.5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "trashpanda-org/gemma-4-31b-larkspur-v0.5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "trashpanda-org/gemma-4-31b-larkspur-v0.5",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/trashpanda-org/gemma-4-31b-larkspur-v0.5

SGLang

How to use trashpanda-org/gemma-4-31b-larkspur-v0.5 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 "trashpanda-org/gemma-4-31b-larkspur-v0.5" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "trashpanda-org/gemma-4-31b-larkspur-v0.5",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "trashpanda-org/gemma-4-31b-larkspur-v0.5" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "trashpanda-org/gemma-4-31b-larkspur-v0.5",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use trashpanda-org/gemma-4-31b-larkspur-v0.5 with Docker Model Runner:
```
docker model run hf.co/trashpanda-org/gemma-4-31b-larkspur-v0.5
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

completely untested, ymmv, but Fuck It We Ball, hopefully it's better at conversation than v0 was.

merged-output

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the TIES merge method using google/gemma-4-31B-it as a base.

Models Merged

The following models were included in the merge:

trashpanda-org/gemma-4-31b-larkspur-v0

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: google/gemma-4-31B-it
  - model: trashpanda-org/gemma-4-31b-larkspur-v0
    parameters:
      density: 1
      weight: 1
merge_method: ties
base_model: google/gemma-4-31B-it
parameters:
  normalize: true
dtype: bfloat16