Instructions to use BananaMind/BananaMind-2-Micro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BananaMind/BananaMind-2-Micro with Transformers:

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

pipe = pipeline("text-generation", model="BananaMind/BananaMind-2-Micro", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("BananaMind/BananaMind-2-Micro", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use BananaMind/BananaMind-2-Micro with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BananaMind/BananaMind-2-Micro"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BananaMind/BananaMind-2-Micro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/BananaMind/BananaMind-2-Micro

SGLang

How to use BananaMind/BananaMind-2-Micro 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 "BananaMind/BananaMind-2-Micro" \
    --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": "BananaMind/BananaMind-2-Micro",
		"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 "BananaMind/BananaMind-2-Micro" \
        --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": "BananaMind/BananaMind-2-Micro",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use BananaMind/BananaMind-2-Micro with Docker Model Runner:
```
docker model run hf.co/BananaMind/BananaMind-2-Micro
```

Final Loss?

by Harley-ml - opened about 21 hours ago

Discussion

Harley-ml

about 21 hours ago

Hey, so I'm wondering, what was the final train or eval loss?

Banaxi-Tech

BananaMind AI org about 11 hours ago

I'll check

Mohammedkarimi

about 10 hours ago

I think the train is a stage that puts pressure on the system and trains the model, and overall it is damaging the system but training the model.

Banaxi-Tech

BananaMind AI org about 10 hours ago

The final training loss is 2.8672. We did not use an eval loss as tiny models are almost entirely unable to overfit on big datasets.

Banaxi-Tech changed discussion status to closed about 10 hours ago

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment