Instructions to use Phonsiri/Thai-Legal-Gemma-4B-CPT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Phonsiri/Thai-Legal-Gemma-4B-CPT with Transformers:

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

pipe = pipeline("text-generation", model="Phonsiri/Thai-Legal-Gemma-4B-CPT")

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("Phonsiri/Thai-Legal-Gemma-4B-CPT")
model = AutoModelForMultimodalLM.from_pretrained("Phonsiri/Thai-Legal-Gemma-4B-CPT", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Phonsiri/Thai-Legal-Gemma-4B-CPT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Phonsiri/Thai-Legal-Gemma-4B-CPT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Phonsiri/Thai-Legal-Gemma-4B-CPT",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Phonsiri/Thai-Legal-Gemma-4B-CPT

SGLang

How to use Phonsiri/Thai-Legal-Gemma-4B-CPT 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 "Phonsiri/Thai-Legal-Gemma-4B-CPT" \
    --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": "Phonsiri/Thai-Legal-Gemma-4B-CPT",
		"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 "Phonsiri/Thai-Legal-Gemma-4B-CPT" \
        --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": "Phonsiri/Thai-Legal-Gemma-4B-CPT",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Phonsiri/Thai-Legal-Gemma-4B-CPT with Docker Model Runner:
```
docker model run hf.co/Phonsiri/Thai-Legal-Gemma-4B-CPT
```

Phonsiri commited on Apr 30

Commit

cc771cd

verified ·

1 Parent(s): 4b37f69

Training in progress, step 5, checkpoint

Browse files

Files changed (4) hide show

checkpoint-5/model.safetensors +1 -1
checkpoint-5/optimizer.pt +2 -2
checkpoint-5/trainer_state.json +10 -10
checkpoint-5/training_args.bin +1 -1

checkpoint-5/model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5fba9d49bb53ed7025504791e27da17b79c2293c7fe8719e2a0379a62e33b8f
 size 15882477500

 version https://git-lfs.github.com/spec/v1
+oid sha256:c89f27c9d1d6030019bbd9bdca903c0058caf0d2d669d52988caae636ca788fd
 size 15882477500

checkpoint-5/optimizer.pt CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93d17a7acbb90d6c10d53eec3e2b6e6066405109fac55e7c25d31bff7e97bfc6
-size 14564634

 version https://git-lfs.github.com/spec/v1
+oid sha256:516bdab405d08a9b56b268d753e0ee6751bb9209aa9923edf1ca32bf6746eff8
+size 29852609572

checkpoint-5/trainer_state.json CHANGED Viewed

@@ -11,37 +11,37 @@
   "log_history": [
     {
       "epoch": 0.010485357362277289,
-      "grad_norm": 3808.0,
       "learning_rate": 0.0,
-      "loss": 1.6072442531585693,
       "step": 1
     },
     {
       "epoch": 0.020970714724554577,
-      "grad_norm": 152.0,
       "learning_rate": 4e-08,
-      "loss": 1.5938644409179688,
       "step": 2
     },
     {
       "epoch": 0.03145607208683186,
-      "grad_norm": 186.0,
       "learning_rate": 8e-08,
-      "loss": 1.593285322189331,
       "step": 3
     },
     {
       "epoch": 0.041941429449109155,
-      "grad_norm": 224.0,
       "learning_rate": 1.2000000000000002e-07,
-      "loss": 1.611860990524292,
       "step": 4
     },
     {
       "epoch": 0.05242678681138644,
-      "grad_norm": 143.0,
       "learning_rate": 1.6e-07,
-      "loss": 1.6161417961120605,
       "step": 5
     }
   ],

   "log_history": [
     {
       "epoch": 0.010485357362277289,
+      "grad_norm": 498.0,
       "learning_rate": 0.0,
+      "loss": 1.6033425331115723,
       "step": 1
     },
     {
       "epoch": 0.020970714724554577,
+      "grad_norm": 139.0,
       "learning_rate": 4e-08,
+      "loss": 1.5843058824539185,
       "step": 2
     },
     {
       "epoch": 0.03145607208683186,
+      "grad_norm": 158.0,
       "learning_rate": 8e-08,
+      "loss": 1.624650239944458,
       "step": 3
     },
     {
       "epoch": 0.041941429449109155,
+      "grad_norm": 201.0,
       "learning_rate": 1.2000000000000002e-07,
+      "loss": 1.5930713415145874,
       "step": 4
     },
     {
       "epoch": 0.05242678681138644,
+      "grad_norm": 1392.0,
       "learning_rate": 1.6e-07,
+      "loss": 1.6284583806991577,
       "step": 5
     }
   ],

checkpoint-5/training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d3874039c01a3b24bfe0ccd8c1eeddac1a45d9a5e1e217213d9e711ad115298
 size 5265

 version https://git-lfs.github.com/spec/v1
+oid sha256:874ac4b635f5e0e0130ccc11de5bfc6cee2eac14aa65567d5f9737b0b5cc556a
 size 5265