Text Generation
Transformers
Safetensors
Catalan
English
Spanish
bloom
finetune
chatml
gpt4
catalan
text-generation-inference
How to use from
SGLangUse 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 "xaviviro/FLOR-6.3B-xat" \
--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": "xaviviro/FLOR-6.3B-xat",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'Quick Links
FLOR-6.3B-xat
FLOR-6.3B-xat és el resultat de finetunejar el model FLOR-6.3B de Projecte Aina amb les instruccions d'OpenAssistant v2 traduïdes automàticament al català amb recursos de Helsinki-NLP i tractades en format ChatML.
Prompt Template
FLOR-6.3B-xat usa ChatML com a prompt template:
<|im_start|>user
Qui va ser Isaac Newton?<|im_end|>
<|im_start|>assistant\n
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 38.23 |
| AI2 Reasoning Challenge (25-Shot) | 38.65 |
| HellaSwag (10-Shot) | 63.76 |
| MMLU (5-Shot) | 26.54 |
| TruthfulQA (0-shot) | 37.96 |
| Winogrande (5-shot) | 62.43 |
| GSM8k (5-shot) | 0.00 |
- Downloads last month
- 9
Model tree for xaviviro/FLOR-6.3B-xat
Base model
projecte-aina/FLOR-6.3B
Install from pip and serve model
# Install SGLang from pip: pip install sglang# Start the SGLang server: python3 -m sglang.launch_server \ --model-path "xaviviro/FLOR-6.3B-xat" \ --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": "xaviviro/FLOR-6.3B-xat", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'