Instructions to use hugging-science/Mochiva-model-research-fix-preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hugging-science/Mochiva-model-research-fix-preview with Transformers:

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

pipe = pipeline("text-generation", model="hugging-science/Mochiva-model-research-fix-preview", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("hugging-science/Mochiva-model-research-fix-preview", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use hugging-science/Mochiva-model-research-fix-preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hugging-science/Mochiva-model-research-fix-preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hugging-science/Mochiva-model-research-fix-preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/hugging-science/Mochiva-model-research-fix-preview

SGLang

How to use hugging-science/Mochiva-model-research-fix-preview 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 "hugging-science/Mochiva-model-research-fix-preview" \
    --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": "hugging-science/Mochiva-model-research-fix-preview",
		"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 "hugging-science/Mochiva-model-research-fix-preview" \
        --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": "hugging-science/Mochiva-model-research-fix-preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use hugging-science/Mochiva-model-research-fix-preview with Docker Model Runner:
```
docker model run hf.co/hugging-science/Mochiva-model-research-fix-preview
```

⭐ Follow Mochiva for updates & better versions

We have started fixing Mochiva! 🎉
This repository is the active repair preview for:

Mochiva-team/Mochiva-model

We are currently working on fixing the model, repairing compatibility issues, and making Mochiva better little by little.
Please be patient while we give this tiny brain some much more fixes.

Transformers support is here! Try our AI Mochi pets here: available soon 🚀
Mochiva by SmilyAI

Mochiva Model For Immersive Roleplay

This model is under active development and is currently being fixed.
It is still a work in progress, but we are actively improving it and working toward a more usable and stable model.

What We’re Doing

We have started fixing the model and are currently:

repairing tokenizer issues
improving model compatibility
checking the weight conversion process
making Mochiva more usable for future updates

Thank you for supporting Mochiva while we work on it!
More updates will come soon as the repair process continues.

GO MOCHIVA!

- SmilyAI Mochiva Team

Downloads last month: 6