Instructions to use Oysiyl/gemma-4-31b-unslop-good-lora-v2-full with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Oysiyl/gemma-4-31b-unslop-good-lora-v2-full with Transformers:

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

pipe = pipeline("image-text-to-text", model="Oysiyl/gemma-4-31b-unslop-good-lora-v2-full")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("Oysiyl/gemma-4-31b-unslop-good-lora-v2-full")
model = AutoModelForMultimodalLM.from_pretrained("Oysiyl/gemma-4-31b-unslop-good-lora-v2-full")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Oysiyl/gemma-4-31b-unslop-good-lora-v2-full with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Oysiyl/gemma-4-31b-unslop-good-lora-v2-full"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Oysiyl/gemma-4-31b-unslop-good-lora-v2-full",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/Oysiyl/gemma-4-31b-unslop-good-lora-v2-full

SGLang

How to use Oysiyl/gemma-4-31b-unslop-good-lora-v2-full 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 "Oysiyl/gemma-4-31b-unslop-good-lora-v2-full" \
    --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": "Oysiyl/gemma-4-31b-unslop-good-lora-v2-full",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "Oysiyl/gemma-4-31b-unslop-good-lora-v2-full" \
        --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": "Oysiyl/gemma-4-31b-unslop-good-lora-v2-full",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Unsloth Studio

How to use Oysiyl/gemma-4-31b-unslop-good-lora-v2-full with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Oysiyl/gemma-4-31b-unslop-good-lora-v2-full to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Oysiyl/gemma-4-31b-unslop-good-lora-v2-full to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Oysiyl/gemma-4-31b-unslop-good-lora-v2-full to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="Oysiyl/gemma-4-31b-unslop-good-lora-v2-full",
    max_seq_length=2048,
)

Docker Model Runner
How to use Oysiyl/gemma-4-31b-unslop-good-lora-v2-full with Docker Model Runner:
```
docker model run hf.co/Oysiyl/gemma-4-31b-unslop-good-lora-v2-full
```

gemma-4-31b-unslop-good-lora-v2-full / training_loss_vs_progress.svg

Oysiyl

Upload training_loss_vs_progress.svg with huggingface_hub

2a483d6 verified 3 months ago

preview code

Raw

History Blame Contribute Delete

6.72 kB

	<svg xmlns="http://www.w3.org/2000/svg" width="1200" height="700" viewBox="0 0 1200 700">
	<rect width="100%" height="100%" fill="#ffffff"/>
	<text x="600.0" y="32" text-anchor="middle" font-family="Arial" font-size="24" fill="#111">Gemma-4-31B: training loss vs normalized progress</text>
	<text x="600.0" y="54" text-anchor="middle" font-family="Arial" font-size="14" fill="#555">Job 69d9faf6cd8c002f31df81c4 • max_steps=250 • points=250</text>
	<line x1="90.00" y1="70" x2="90.00" y2="610" stroke="#eee"/>
	<text x="90.00" y="635" text-anchor="middle" font-family="Arial" font-size="12" fill="#555">0%</text>
	<line x1="304.00" y1="70" x2="304.00" y2="610" stroke="#eee"/>
	<text x="304.00" y="635" text-anchor="middle" font-family="Arial" font-size="12" fill="#555">20%</text>
	<line x1="518.00" y1="70" x2="518.00" y2="610" stroke="#eee"/>
	<text x="518.00" y="635" text-anchor="middle" font-family="Arial" font-size="12" fill="#555">40%</text>
	<line x1="732.00" y1="70" x2="732.00" y2="610" stroke="#eee"/>
	<text x="732.00" y="635" text-anchor="middle" font-family="Arial" font-size="12" fill="#555">60%</text>
	<line x1="946.00" y1="70" x2="946.00" y2="610" stroke="#eee"/>
	<text x="946.00" y="635" text-anchor="middle" font-family="Arial" font-size="12" fill="#555">80%</text>
	<line x1="1160.00" y1="70" x2="1160.00" y2="610" stroke="#eee"/>
	<text x="1160.00" y="635" text-anchor="middle" font-family="Arial" font-size="12" fill="#555">100%</text>
	<line x1="90" y1="610.00" x2="1160" y2="610.00" stroke="#eee"/>
	<text x="80" y="614.00" text-anchor="end" font-family="Arial" font-size="12" fill="#555">1.16</text>
	<line x1="90" y1="502.00" x2="1160" y2="502.00" stroke="#eee"/>
	<text x="80" y="506.00" text-anchor="end" font-family="Arial" font-size="12" fill="#555">1.71</text>
	<line x1="90" y1="394.00" x2="1160" y2="394.00" stroke="#eee"/>
	<text x="80" y="398.00" text-anchor="end" font-family="Arial" font-size="12" fill="#555">2.27</text>
	<line x1="90" y1="286.00" x2="1160" y2="286.00" stroke="#eee"/>
	<text x="80" y="290.00" text-anchor="end" font-family="Arial" font-size="12" fill="#555">2.83</text>
	<line x1="90" y1="178.00" x2="1160" y2="178.00" stroke="#eee"/>
	<text x="80" y="182.00" text-anchor="end" font-family="Arial" font-size="12" fill="#555">3.38</text>
	<line x1="90" y1="70.00" x2="1160" y2="70.00" stroke="#eee"/>
	<text x="80" y="74.00" text-anchor="end" font-family="Arial" font-size="12" fill="#555">3.94</text>
	<rect x="90" y="70" width="1070" height="540" fill="none" stroke="#bbb"/>
	<polyline fill="none" stroke="#2563eb" stroke-width="2" points="94.28,249.15 98.56,209.15 102.84,197.63 107.12,107.24 111.40,204.78 115.68,253.26 119.96,140.75 124.24,213.42 128.52,234.20 132.80,324.12 137.08,394.41 141.36,399.09 145.64,345.41 149.92,416.61 154.20,397.78 158.48,345.48 162.76,384.95 167.04,330.45 171.32,426.49 175.60,450.46 179.88,416.61 184.16,449.05 188.44,413.93 192.72,392.22 197.00,411.56 201.28,393.51 205.56,439.91 209.84,475.32 214.12,453.36 218.40,449.04 222.68,436.30 226.96,411.32 231.24,432.06 235.52,495.59 239.80,386.38 244.08,444.75 248.36,451.50 252.64,414.97 256.92,442.93 261.20,469.22 265.48,442.15 269.76,428.52 274.04,497.88 278.32,443.64 282.60,440.34 286.88,403.93 291.16,458.18 295.44,422.18 299.72,458.90 304.00,408.54 308.28,438.69 312.56,498.60 316.84,459.84 321.12,451.02 325.40,478.57 329.68,457.00 333.96,483.58 338.24,443.72 342.52,453.62 346.80,430.88 351.08,420.82 355.36,402.98 359.64,447.61 363.92,450.37 368.20,462.94 372.48,488.85 376.76,494.45 381.04,437.02 385.32,463.02 389.60,446.12 393.88,457.52 398.16,468.95 402.44,492.29 406.72,484.67 411.00,506.09 415.28,435.42 419.56,492.51 423.84,423.17 428.12,428.52 432.40,490.05 436.68,480.72 440.96,490.30 445.24,468.61 449.52,450.45 453.80,461.81 458.08,437.61 462.36,443.21 466.64,518.05 470.92,470.90 475.20,497.14 479.48,458.85 483.76,482.27 488.04,463.47 492.32,525.69 496.60,464.51 500.88,572.76 505.16,502.43 509.44,505.93 513.72,505.55 518.00,502.34 522.28,446.29 526.56,478.18 530.84,446.84 535.12,487.09 539.40,490.69 543.68,505.63 547.96,466.01 552.24,448.32 556.52,455.33 560.80,498.23 565.08,469.33 569.36,459.77 573.64,476.17 577.92,409.52 582.20,458.80 586.48,497.33 590.76,468.11 595.04,488.64 599.32,483.85 603.60,495.00 607.88,456.05 612.16,451.15 616.44,488.72 620.72,513.93 625.00,458.87 629.28,457.89 633.56,500.98 637.84,472.86 642.12,497.94 646.40,516.04 650.68,465.75 654.96,479.36 659.24,454.93 663.52,478.24 667.80,477.45 672.08,463.20 676.36,466.96 680.64,444.29 684.92,494.89 689.20,457.07 693.48,503.17 697.76,492.87 702.04,474.82 706.32,483.85 710.60,535.89 714.88,445.81 719.16,439.70 723.44,532.72 727.72,432.33 732.00,450.57 736.28,450.46 740.56,521.26 744.84,468.32 749.12,487.66 753.40,524.64 757.68,440.13 761.96,455.62 766.24,531.44 770.52,491.31 774.80,498.58 779.08,463.49 783.36,519.01 787.64,433.30 791.92,435.34 796.20,468.40 800.48,474.18 804.76,496.17 809.04,467.99 813.32,474.97 817.60,413.37 821.88,466.00 826.16,478.30 830.44,547.33 834.72,526.35 839.00,480.19 843.28,536.04 847.56,465.73 851.84,504.53 856.12,515.96 860.40,474.48 864.68,465.16 868.96,465.37 873.24,520.80 877.52,443.07 881.80,495.53 886.08,505.03 890.36,516.55 894.64,488.58 898.92,520.76 903.20,446.31 907.48,525.47 911.76,455.49 916.04,483.56 920.32,435.25 924.60,484.94 928.88,531.56 933.16,488.10 937.44,495.69 941.72,474.24 946.00,451.45 950.28,472.23 954.56,512.40 958.84,485.32 963.12,466.21 967.40,477.13 971.68,482.88 975.96,549.96 980.24,429.37 984.52,465.74 988.80,479.81 993.08,427.91 997.36,509.61 1001.64,438.44 1005.92,517.84 1010.20,474.00 1014.48,477.72 1018.76,423.01 1023.04,532.99 1027.32,496.48 1031.60,478.81 1035.88,518.60 1040.16,483.83 1044.44,502.07 1048.72,430.88 1053.00,506.82 1057.28,518.94 1061.56,450.96 1065.84,524.27 1070.12,522.24 1074.40,535.38 1078.68,441.81 1082.96,550.36 1087.24,552.12 1091.52,478.46 1095.80,477.27 1100.08,453.83 1104.36,476.19 1108.64,504.38 1112.92,488.91 1117.20,471.17 1121.48,490.17 1125.76,464.16 1130.04,447.97 1134.32,489.76 1138.60,486.76 1142.88,525.58 1147.16,433.26 1151.44,492.80 1155.72,517.49 1160.00,442.49"/>
	<circle cx="94.28" cy="249.15" r="3.5" fill="#059669"/>
	<circle cx="1160.00" cy="442.49" r="3.5" fill="#dc2626"/>
	<text x="102.28" y="241.15" font-family="Arial" font-size="12" fill="#059669">first 3.017</text>
	<text x="1152.00" y="434.49" text-anchor="end" font-family="Arial" font-size="12" fill="#dc2626">last 2.020</text>
	<text x="600.0" y="675" text-anchor="middle" font-family="Arial" font-size="16" fill="#333">normalized progress (step / max_steps)</text>
	<text transform="translate(24 350.0) rotate(-90)" text-anchor="middle" font-family="Arial" font-size="16" fill="#333">training loss</text>
	</svg>