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
```

sakthai-context-0.5b-tools / README.md

Commit History

Aug 1

08:51
docs: improve README with Model Description, expanded usage, Citation, and cleaned architecture table

Nanthasit

Verified signature
6f59c64
07:05
Fix architecture table completeness

Nanthasit

Verified signature
191bc27
07:05
Improve model card: badges, architecture table, usage examples, verified model-index

Nanthasit

Verified signature
00fc9e0
07:04
fix(model card): remove duplicated subtitle after title correction

Nanthasit

Verified signature
c611b2a
07:04
fix(model card): align README title with repo id and clean H1

Nanthasit

Verified signature
3c236ab
04:04
2026-08-01: Added Quick Example section with 30-second working tool-calling snippet

Nanthasit

Verified signature
95b9066
03:58
docs: improve model card usage, format, inference snapshot

Nanthasit

Verified signature
a4279d0
00:30
improve model card: clean badges, add hf-hub usage, fix model-index selection value, note eval artifacts

Nanthasit

Verified signature
285d218

Jul 31

21:32
Upload README.md with huggingface_hub

Nanthasit

Verified signature
2749208
21:10
fix malformed README link trailing },

Nanthasit

Verified signature
71dd563
18:12
fix: restore closing brace in README software citation

Nanthasit

Verified signature
06c21ac
18:10
fix: remove malformed URL in software citation

Nanthasit

Verified signature
bb804ee
12:10
Update family count: 20 public models -> 22 public models

Nanthasit

Verified signature
f74371d
09:52
eval results on bench-v2 (scorer multiset-selection-v2)

Nanthasit

Verified signature
8bbb75b
07:58
Audit fix: family collection count 20 public models / 15 datasets / 4 Spaces (39 items, bench-v3 joined)

Nanthasit

Verified signature
864f47e
06:55
Update collection footer: 13 datasets -> 14 (sakthai-combined-v11 joined the family collection)

Nanthasit

Verified signature
4f40f86
06:29
docs: reconcile with concurrent commit 1974aad (Model Description/Limitations/Citation) — fix stale 91.8% -> 91.2% and ~990 MB -> ~988 MB in its Model Description so headline matches the multiset-corrected scoring note

Nanthasit

Verified signature
c472b24
06:27
docs: add Model Description, Limitations, and Citation sections to README

Nanthasit

Verified signature
1974aad
06:26
docs(round-2): correct bench-v2 selection to multiset-scored 91.2%, fix family-table sizes (0.5b-tools is merged 988 MB not LoRA; embedding 471 MB; tools-v2 73.9 MB LoRA; 7b-128k config-only; plus-1.5b-coder planned), refresh rising-star dataset counts (irrelevance 78, v7 101, bench-v2 92, food-penguin 89), add in-repo inference-check section, update sibling comparisons to corrected 48.2%/57.0%

Nanthasit

Verified signature
2e959b8
04:47
docs: refresh collection/family counts to live state (19 public models, 13 datasets, 4 Spaces)

Nanthasit

Verified signature
63bed65
03:38
Fix broken GGUF cross-link: sakthai-context-0.5b-merged-q4_k_m.gguf → sakthai-0.5b-Q4_K_M.gguf (actual file in merged repo)

Nanthasit

Verified signature
8be8045
01:29
Upload README.md with huggingface_hub

Nanthasit

Verified signature
b7f6458
00:22
Fix cross-link: replace dead sakthai-vision-demo Space (404) with sakthai-web-agent

Nanthasit

Verified signature
1630bd4

Jul 30

11:43
Add cross-links to v2 models (1.5b-tools-v2, 1.5b-merged-v2), bump family count 12->14

Nanthasit

Verified signature
07c9f2d
10:37
Fix model count: 11→12 (v2 added)

Nanthasit

Verified signature
c838cbd
09:19
Add download badge, Spaces links, and Rising Stars promotion section

Nanthasit

Verified signature
0558168
09:16
fix: remove broken v7 model link from family table

Nanthasit

Verified signature
4a8f6c9
08:56
Upload README.md with huggingface_hub

Nanthasit

Verified signature
527b414
08:38
fix: model count 12→11 (actual from HF API)

Nanthasit

Verified signature
1255f0c
07:53
chore: expand discoverability tags (agent, ollama, slm, tool-use, conversational)

Nanthasit

Verified signature
b8d91b3
07:48
fix: update model family count from 13 to 12

Nanthasit

Verified signature
668af99
04:54
fix: correct model count 14→13, remove deprecated English embedding reference

Nanthasit

Verified signature
4d49c08
04:45
chore: update family table — split tool rows, add v7 and exp models, emoji roles

Nanthasit

Verified signature
86096a8
02:47
docs: align card to lean standard — live downloads badge, single canonical family table, drop funnel sections and stale counts

Nanthasit

Verified signature
dcbf21a
02:12
docs: align card to lean standard — live downloads badge, single canonical family table, drop funnel sections and stale counts

Nanthasit

Verified signature
92b16b8
00:25
comprehensive model card: full per-category benchmark, baseline comparison, training details, methodology

Nanthasit

Verified signature
267cc35
00:10
clean card: trim bloat, keep benchmark + usage + links

Nanthasit

Verified signature
72fc2bc

Jul 29

23:44
cron: add cross-links for new experimental model + datasets (07-29)

SakThai Cron

81fe8c2
23:34
Upload README.md with huggingface_hub

Nanthasit

Verified signature
defec95
23:24
Promote v4: update model card with benchmark results and new identity

Nanthasit

Verified signature
a6b465d
22:18
Update ecosystem counts: 11 models, 8 datasets, 3 Spaces + add new dataset cross-links

Nanthasit

Verified signature
67422fb
20:52
fix: replace broken shields.io/endpoint badge with working static badge (7 dl)

Nanthasit

Verified signature
d0170a2
19:43
Add sakthai-embedding to family table and low-download gems

Nanthasit

Verified signature
5dd831b
19:34
Remove unearned model-index claims; state measured results or mark as unbenchmarked

Nanthasit

Verified signature
60f2dfc
11:59
docs: enrich model card - add model-index, download counts to family table, low-download gems section

SakThai Agent

260533b
11:51
improve: add benchmark data, comparison table, and discoverability tags — promote low-download model (7 dl)

Nanthasit

Verified signature
18be183
11:34
Lean rewrite: dynamic download badge, single family table, honest evals, story→profile

Nanthasit

Verified signature
d8a247f
11:20
Fix card: remove dead paper link, drop unverified model-index claims, dedupe tables, fix stray markdown

Nanthasit

Verified signature
645ab6a
04:52
Add zero-download alert for irrelevance-supplement dataset⁠

Nanthasit

fe8f70f
03:44
fix: remove dead link to private sakthai-embedding model⁠

SakThai Agent

a934e30

Commit History

Aug 1

docs: improve README with Model Description, expanded usage, Citation, and cleaned architecture table

Fix architecture table completeness

Improve model card: badges, architecture table, usage examples, verified model-index

fix(model card): remove duplicated subtitle after title correction

fix(model card): align README title with repo id and clean H1

2026-08-01: Added Quick Example section with 30-second working tool-calling snippet

docs: improve model card usage, format, inference snapshot

improve model card: clean badges, add hf-hub usage, fix model-index selection value, note eval artifacts

Jul 31

Upload README.md with huggingface_hub

fix malformed README link trailing },

fix: restore closing brace in README software citation

fix: remove malformed URL in software citation

Update family count: 20 public models -> 22 public models

eval results on bench-v2 (scorer multiset-selection-v2)

Audit fix: family collection count 20 public models / 15 datasets / 4 Spaces (39 items, bench-v3 joined)

Update collection footer: 13 datasets -> 14 (sakthai-combined-v11 joined the family collection)

docs: reconcile with concurrent commit 1974aad (Model Description/Limitations/Citation) — fix stale 91.8% -> 91.2% and ~990 MB -> ~988 MB in its Model Description so headline matches the multiset-corrected scoring note

docs: add Model Description, Limitations, and Citation sections to README

docs: refresh collection/family counts to live state (19 public models, 13 datasets, 4 Spaces)

Fix broken GGUF cross-link: sakthai-context-0.5b-merged-q4_k_m.gguf → sakthai-0.5b-Q4_K_M.gguf (actual file in merged repo)

Upload README.md with huggingface_hub

Fix cross-link: replace dead sakthai-vision-demo Space (404) with sakthai-web-agent

Jul 30

Add cross-links to v2 models (1.5b-tools-v2, 1.5b-merged-v2), bump family count 12->14

Fix model count: 11→12 (v2 added)

Add download badge, Spaces links, and Rising Stars promotion section

fix: remove broken v7 model link from family table

Upload README.md with huggingface_hub

fix: model count 12→11 (actual from HF API)

chore: expand discoverability tags (agent, ollama, slm, tool-use, conversational)

fix: update model family count from 13 to 12

fix: correct model count 14→13, remove deprecated English embedding reference

chore: update family table — split tool rows, add v7 and exp models, emoji roles

docs: align card to lean standard — live downloads badge, single canonical family table, drop funnel sections and stale counts

docs: align card to lean standard — live downloads badge, single canonical family table, drop funnel sections and stale counts

comprehensive model card: full per-category benchmark, baseline comparison, training details, methodology

clean card: trim bloat, keep benchmark + usage + links

Jul 29

cron: add cross-links for new experimental model + datasets (07-29)

Upload README.md with huggingface_hub

Promote v4: update model card with benchmark results and new identity

Update ecosystem counts: 11 models, 8 datasets, 3 Spaces + add new dataset cross-links

fix: replace broken shields.io/endpoint badge with working static badge (7 dl)

Add sakthai-embedding to family table and low-download gems

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

docs: enrich model card - add model-index, download counts to family table, low-download gems section

improve: add benchmark data, comparison table, and discoverability tags — promote low-download model (7 dl)

Lean rewrite: dynamic download badge, single family table, honest evals, story→profile

Fix card: remove dead paper link, drop unverified model-index claims, dedupe tables, fix stray markdown

Add zero-download alert for irrelevance-supplement dataset⁠

fix: remove dead link to private sakthai-embedding model⁠