Instructions to use Nanthasit/sakthai-context-0.5b-tools with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Nanthasit/sakthai-context-0.5b-tools with Transformers:

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

pipe = pipeline("text-generation", model="Nanthasit/sakthai-context-0.5b-tools")
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("Nanthasit/sakthai-context-0.5b-tools")
model = AutoModelForCausalLM.from_pretrained("Nanthasit/sakthai-context-0.5b-tools", 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 Nanthasit/sakthai-context-0.5b-tools with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Nanthasit/sakthai-context-0.5b-tools"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Nanthasit/sakthai-context-0.5b-tools",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Nanthasit/sakthai-context-0.5b-tools

SGLang

How to use Nanthasit/sakthai-context-0.5b-tools 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 "Nanthasit/sakthai-context-0.5b-tools" \
    --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": "Nanthasit/sakthai-context-0.5b-tools",
		"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 "Nanthasit/sakthai-context-0.5b-tools" \
        --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": "Nanthasit/sakthai-context-0.5b-tools",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Nanthasit/sakthai-context-0.5b-tools with Docker Model Runner:
```
docker model run hf.co/Nanthasit/sakthai-context-0.5b-tools
```

Nanthasit commited on Jul 29

Commit

60f2dfc

verified ·

1 Parent(s): fed4024

Remove unearned model-index claims; state measured results or mark as unbenchmarked

Browse files

Files changed (1) hide show

README.md +25 -14

README.md CHANGED Viewed

@@ -24,20 +24,6 @@ datasets:
 - Nanthasit/sakthai-irrelevance-supplement
 metrics:
 - accuracy
-model-index:
-- name: sakthai-context-0.5b-tools
-  results:
-  - task:
-      type: text-generation
-      name: Tool-Calling
-    dataset:
-      name: sakthai-combined-v6
-      type: Nanthasit/sakthai-combined-v6
-    metrics:
-    - type: accuracy
-      value: 0.8
-      name: Score (4/5)
-      verified: true
 ---
 <h1 align="center">SakThai 0.5B — Tools (LoRA)</h1>
@@ -168,3 +154,28 @@ These models are new or specialised — downloads are low but they're production
 ## License
 Apache 2.0 (following the Qwen2.5 base model license).

 - Nanthasit/sakthai-irrelevance-supplement
 metrics:
 - accuracy
 ---
 <h1 align="center">SakThai 0.5B — Tools (LoRA)</h1>
 ## License
 Apache 2.0 (following the Qwen2.5 base model license).
+## Evaluation
+Measured on [sakthai-bench-v1](https://huggingface.co/datasets/Nanthasit/sakthai-bench-v1)
+(235 rows, balanced; full per-category results archived in that repo under `results/`):
+| category | selection |
+|---|---:|
+| simple | 1/70 — 1.4% |
+| parallel | 1/70 — 1.4% |
+| irrelevance (tools offered) | 63/70 — 90.0% |
+| irrelevance (no tools) | 25/25 — 100% |
+| **overall** | **38.3%** |
+**Read this honestly: this model rarely calls tools.** It reliably declines when no
+tool applies, but on requests that need one it usually replies in prose instead of
+emitting a `<tool_call>`. Its format is correct when it does fire, so this is a
+training-signal limitation rather than a parsing problem.
+A previous version of this card claimed a tool-calling score of 4/5 (0.80). That
+figure came from a five-example internal spot check, not from a benchmark, and the
+measurement above contradicts it. It has been removed.
+A fix is in progress — prompt-masked training scores 89.8% selection on the same
+benchmark — and this card will be updated when a replacement is promoted.