Instructions to use LLM-OS-Models/gdn1-16k32k-bridge-1b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LLM-OS-Models/gdn1-16k32k-bridge-1b with Transformers:

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

pipe = pipeline("text-generation", model="LLM-OS-Models/gdn1-16k32k-bridge-1b", device_map="auto")

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("LLM-OS-Models/gdn1-16k32k-bridge-1b", dtype="auto", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use LLM-OS-Models/gdn1-16k32k-bridge-1b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LLM-OS-Models/gdn1-16k32k-bridge-1b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM-OS-Models/gdn1-16k32k-bridge-1b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/LLM-OS-Models/gdn1-16k32k-bridge-1b

SGLang

How to use LLM-OS-Models/gdn1-16k32k-bridge-1b 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 "LLM-OS-Models/gdn1-16k32k-bridge-1b" \
    --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": "LLM-OS-Models/gdn1-16k32k-bridge-1b",
		"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 "LLM-OS-Models/gdn1-16k32k-bridge-1b" \
        --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": "LLM-OS-Models/gdn1-16k32k-bridge-1b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use LLM-OS-Models/gdn1-16k32k-bridge-1b with Docker Model Runner:
```
docker model run hf.co/LLM-OS-Models/gdn1-16k32k-bridge-1b
```

gdn1-16k32k-bridge-1b / train_plan.json

gyung

Upload GDN1 16K/32K Bridge 1B Full Fine-Tune

682e4c0 verified 26 days ago

Raw

History Blame Contribute Delete

1.13 kB

	{
	"mode": "full_finetune",
	"target_hardware": "8x NVIDIA H200",
	"model_path": "/home/work/.projects/LLM-OS-Models/long-gdn/runs/gdn1_32k_balanced_recovery_1b_bs10_ft/checkpoint-200",
	"tokenizer_path": "/home/work/.projects/LLM-OS-Models/long-gdn/runs/gdn1_32k_balanced_recovery_1b_bs10_ft/final",
	"manifest": "/home/work/.projects/LLM-OS-Models/long-gdn/configs/gdn1_memory_mix_16k32k_bridge_recovery.json",
	"output_dir": "/home/work/.projects/LLM-OS-Models/long-gdn/runs/gdn1_16k32k_bridge_from_balanced200_1b_bs10_ft",
	"seq_len": 32768,
	"per_device_batch_size": 10,
	"grad_accum": 1,
	"world_size_assumed": 8,
	"global_tokens_per_step": 2621440,
	"max_steps": 382,
	"target_tokens": 1001390080,
	"learning_rate": 5e-06,
	"warmup_steps": 25,
	"weight_decay": 0.1,
	"optim": "adamw_torch_fused",
	"bf16": true,
	"gradient_checkpointing": true,
	"override_max_position_embeddings": null,
	"notes": [
	"No LoRA/PEFT adapters are used.",
	"All model parameters remain trainable unless the model implementation freezes them.",
	"Dry-run writes this plan without loading model weights."
	]
	}