Instructions to use FrontiersMind/Nandi-Mini-600M-Early-Checkpoint with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FrontiersMind/Nandi-Mini-600M-Early-Checkpoint with Transformers:

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

pipe = pipeline("text-generation", model="FrontiersMind/Nandi-Mini-600M-Early-Checkpoint", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("FrontiersMind/Nandi-Mini-600M-Early-Checkpoint", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use FrontiersMind/Nandi-Mini-600M-Early-Checkpoint with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "FrontiersMind/Nandi-Mini-600M-Early-Checkpoint"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "FrontiersMind/Nandi-Mini-600M-Early-Checkpoint",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/FrontiersMind/Nandi-Mini-600M-Early-Checkpoint

SGLang

How to use FrontiersMind/Nandi-Mini-600M-Early-Checkpoint 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 "FrontiersMind/Nandi-Mini-600M-Early-Checkpoint" \
    --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": "FrontiersMind/Nandi-Mini-600M-Early-Checkpoint",
		"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 "FrontiersMind/Nandi-Mini-600M-Early-Checkpoint" \
        --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": "FrontiersMind/Nandi-Mini-600M-Early-Checkpoint",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use FrontiersMind/Nandi-Mini-600M-Early-Checkpoint with Docker Model Runner:
```
docker model run hf.co/FrontiersMind/Nandi-Mini-600M-Early-Checkpoint
```

aksinghyaani commited on 8 days ago

Commit

b6aee65

verified ·

1 Parent(s): cff872c

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -36,7 +36,7 @@ Nandi-Mini-600M introduces several efficiency-focused architectural optimization
 #### Shared KV (Shared Key-Value Vectors)
-Shared KV is one of the core architectural ideas explored in Nandi-Mini. Instead of storing separate Key and Value vectors, both share the same underlying representation, while a lightweight Key normalization step is applied specifically for attention computation.
 This design reduces KV-cache memory usage by ~50% during inference with only a small increase in compute overhead, since RoPE and Key normalization are applied dynamically during attention computation.

 #### Shared KV (Shared Key-Value Vectors)
+Shared KV is one of the core architectural ideas explored in Nandi-Mini. Instead of computing separate Key and Value projections, both reuse a shared latent representation, while a lightweight Key normalization step is applied specifically for attention computation.
 This design reduces KV-cache memory usage by ~50% during inference with only a small increase in compute overhead, since RoPE and Key normalization are applied dynamically during attention computation.