Instructions to use iamshnoo/combined_without_metadata_3b_chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use iamshnoo/combined_without_metadata_3b_chat with Transformers:

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

pipe = pipeline("text-generation", model="iamshnoo/combined_without_metadata_3b_chat")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("iamshnoo/combined_without_metadata_3b_chat")
model = AutoModelForCausalLM.from_pretrained("iamshnoo/combined_without_metadata_3b_chat", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use iamshnoo/combined_without_metadata_3b_chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "iamshnoo/combined_without_metadata_3b_chat"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "iamshnoo/combined_without_metadata_3b_chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/iamshnoo/combined_without_metadata_3b_chat

SGLang

How to use iamshnoo/combined_without_metadata_3b_chat 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 "iamshnoo/combined_without_metadata_3b_chat" \
    --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": "iamshnoo/combined_without_metadata_3b_chat",
		"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 "iamshnoo/combined_without_metadata_3b_chat" \
        --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": "iamshnoo/combined_without_metadata_3b_chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use iamshnoo/combined_without_metadata_3b_chat with Docker Model Runner:
```
docker model run hf.co/iamshnoo/combined_without_metadata_3b_chat
```

iamshnoo commited on Apr 2

Commit

f42a00f

verified ·

1 Parent(s): fc28306

Update model card and embedded training curves

Browse files

Files changed (4) hide show

README.md +18 -2
assets/grad_norm.png +0 -0
assets/learning_rate.png +0 -0
assets/train_loss.png +0 -0

README.md CHANGED Viewed

@@ -15,7 +15,7 @@ tags:
 ## Summary
-This repo contains the merged chat model for the combined without metadata 3b branch of the metadata localization project. It was produced by supervised fine-tuning on the project QA benchmark after continued pretraining.
 ## Variant Metadata
@@ -60,6 +60,22 @@ This repo contains the merged chat model for the combined without metadata 3b br
 - `per_device_train_batch_size=2`, `gradient_accumulation_steps=8`
 - LoRA targets: `q_proj`, `k_proj`, `v_proj`, `o_proj`, `gate_proj`, `up_proj`, `down_proj`
 ## Project Context
 This model is part of the metadata localization release. Related checkpoints and variants are grouped in the public Hugging Face collection [Metadata Conditioned LLMs](https://huggingface.co/collections/iamshnoo/metadata-conditioned-llms).
@@ -67,4 +83,4 @@ This model is part of the metadata localization release. Related checkpoints and
 - Project repository: [https://github.com/iamshnoo/metadata_localization](https://github.com/iamshnoo/metadata_localization)
 - Paper: [https://arxiv.org/abs/2601.15236](https://arxiv.org/abs/2601.15236)
-Last synced: `2026-04-02 13:51:15 UTC`

 ## Summary
+This repo contains the merged chat model for the combined without metadata 3b branch of the metadata localization project. It was produced by supervised fine-tuning on the project QA benchmark after project pretraining.
 ## Variant Metadata
 - `per_device_train_batch_size=2`, `gradient_accumulation_steps=8`
 - LoRA targets: `q_proj`, `k_proj`, `v_proj`, `o_proj`, `gate_proj`, `up_proj`, `down_proj`
+## Training Curves
+Static plots below were exported from the private Weights & Biases run and embedded here for public access.
+### Train Loss
+![Train Loss](assets/train_loss.png)
+### Learning Rate
+![Learning Rate](assets/learning_rate.png)
+### Gradient Norm
+![Gradient Norm](assets/grad_norm.png)
 ## Project Context
 This model is part of the metadata localization release. Related checkpoints and variants are grouped in the public Hugging Face collection [Metadata Conditioned LLMs](https://huggingface.co/collections/iamshnoo/metadata-conditioned-llms).
 - Project repository: [https://github.com/iamshnoo/metadata_localization](https://github.com/iamshnoo/metadata_localization)
 - Paper: [https://arxiv.org/abs/2601.15236](https://arxiv.org/abs/2601.15236)
+Last synced: `2026-04-02 14:48:10 UTC`

assets/grad_norm.png ADDED Viewed

assets/learning_rate.png ADDED Viewed

assets/train_loss.png ADDED Viewed