Instructions to use cockroach54/solar-sft-qlora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cockroach54/solar-sft-qlora with Transformers:

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

pipe = pipeline("text-generation", model="cockroach54/solar-sft-qlora")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cockroach54/solar-sft-qlora")
model = AutoModelForCausalLM.from_pretrained("cockroach54/solar-sft-qlora", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cockroach54/solar-sft-qlora with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cockroach54/solar-sft-qlora"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cockroach54/solar-sft-qlora",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/cockroach54/solar-sft-qlora

SGLang

How to use cockroach54/solar-sft-qlora 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 "cockroach54/solar-sft-qlora" \
    --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": "cockroach54/solar-sft-qlora",
		"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 "cockroach54/solar-sft-qlora" \
        --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": "cockroach54/solar-sft-qlora",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use cockroach54/solar-sft-qlora with Docker Model Runner:
```
docker model run hf.co/cockroach54/solar-sft-qlora
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

just for practice
trained by transformers, peft, bitsandbytes
evaluated by https://github.com/davidkim205/kollm_evaluation

Tasks	Version	Filter	Metric	Value		Stderr
kobest_wic	1	none	acc	0.5056	±	0.0141
		none	f1	0.4170	±	N/A
kobest_sentineg	1	none	acc	0.8791	±	0.0164
		none	f1	0.8779	±	N/A
kobest_hellaswag	1	none	acc	0.4540	±	0.0223
		none	f1	0.4522	±	N/A
		none	acc_norm	0.5600	±	0.0222
kobest_copa	1	none	acc	0.6540	±	0.0151
		none	f1	0.6535	±	N/A
kobest_boolq	1	none	acc	0.5135	±	0.0133
		none	f1	0.3592	±	N/A
ko_truthfulqa	2	none	acc	0.2938	±	0.0159
ko_hellaswag	1	none	acc	0.3816	±	0.0048
		none	acc_norm	0.5171	±	0.0050
ko_common_gen	1	none	acc	0.8095	±	0.0100
		none	acc_norm	0.8095	±	0.0100
ko_arc_easy	1	none	acc	0.3311	±	0.0138
		none	acc_norm	0.4053	±	0.0143

Downloads last month: 29

Safetensors

Model size

11B params

Tensor type

F16

Model tree for cockroach54/solar-sft-qlora

Quantizations

4 models

cockroach54
/

solar-sft-qlora

Model tree for cockroach54/solar-sft-qlora

Dataset used to train cockroach54/solar-sft-qlora

Spaces using cockroach54/solar-sft-qlora 8