Instructions to use budecosystem/hex-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use budecosystem/hex-1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="budecosystem/hex-1") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("budecosystem/hex-1") model = AutoModelForCausalLM.from_pretrained("budecosystem/hex-1", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use budecosystem/hex-1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "budecosystem/hex-1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "budecosystem/hex-1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/budecosystem/hex-1
- SGLang
How to use budecosystem/hex-1 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 "budecosystem/hex-1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "budecosystem/hex-1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'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 "budecosystem/hex-1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "budecosystem/hex-1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use budecosystem/hex-1 with Docker Model Runner:
docker model run hf.co/budecosystem/hex-1
Update README.md
Browse files
README.md
CHANGED
|
@@ -25,17 +25,8 @@ This model is a fine-tuned version of [Qwen/Qwen3-4B-Base](https://huggingface.c
|
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
## Intended uses & limitations
|
| 31 |
-
|
| 32 |
-
More information needed
|
| 33 |
-
|
| 34 |
-
## Training and evaluation data
|
| 35 |
-
|
| 36 |
-
More information needed
|
| 37 |
-
|
| 38 |
-
## Training procedure
|
| 39 |
|
| 40 |
### Training hyperparameters
|
| 41 |
|
|
@@ -55,6 +46,16 @@ The following hyperparameters were used during training:
|
|
| 55 |
|
| 56 |
### Training results
|
| 57 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
|
| 59 |
|
| 60 |
### Framework versions
|
|
|
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
| 28 |
+
Hex-1 is a 4-billion parameter language model specifically optimized for Indian languages. It supports five major Indian languages, including Hindi, Kannada, Telugu, Tamil and Malayalam.
|
| 29 |
+
When benchmarked against leading models like Gemma-2B, LLaMA-3.2-3B, and Sarvam-1, Hex1 delivers best-in-class performance in all five supported languages on MMLU dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
|
| 31 |
### Training hyperparameters
|
| 32 |
|
|
|
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
+
### Performance Comparison on ARCC Dataset
|
| 50 |
+
|
| 51 |
+
| Benchmark | Gemma-2-2B | Llama-3.2-3B | Llama-3.1-8B | Sarvam-1 | Hex-1 |
|
| 52 |
+
|-----------|------------|--------------|---------------|-----------|--------|
|
| 53 |
+
| arcc_hi | 37.57 | 49.13 | 56.17 | 60.00 | 36.68 |
|
| 54 |
+
| arcc_ta | 32.78 | 34.70 | 44.78 | 57.04 | 38.65 |
|
| 55 |
+
| arcc_te | 30.00 | 34.09 | 43.04 | 59.39 | 37.96 |
|
| 56 |
+
| arcc_kn | 29.22 | 36.43 | 44.70 | 57.04 | 38.31 |
|
| 57 |
+
| arcc_ml | 29.91 | 33.22 | 46.78 | 58.96 | 29.60 |
|
| 58 |
+
|
| 59 |
|
| 60 |
|
| 61 |
### Framework versions
|