Instructions to use palette-lab/songgot-l with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use palette-lab/songgot-l with Transformers:

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

pipe = pipeline("text-generation", model="palette-lab/songgot-l")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("palette-lab/songgot-l")
model = AutoModelForCausalLM.from_pretrained("palette-lab/songgot-l", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use palette-lab/songgot-l with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf palette-lab/songgot-l:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf palette-lab/songgot-l:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf palette-lab/songgot-l:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf palette-lab/songgot-l:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf palette-lab/songgot-l:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf palette-lab/songgot-l:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf palette-lab/songgot-l:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf palette-lab/songgot-l:Q4_K_M

Use Docker

docker model run hf.co/palette-lab/songgot-l:Q4_K_M

LM Studio
Jan

vLLM

How to use palette-lab/songgot-l with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/palette-lab/songgot-l:Q4_K_M

SGLang

How to use palette-lab/songgot-l 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 "palette-lab/songgot-l" \
    --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": "palette-lab/songgot-l",
		"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 "palette-lab/songgot-l" \
        --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": "palette-lab/songgot-l",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use palette-lab/songgot-l with Ollama:
```
ollama run hf.co/palette-lab/songgot-l:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use palette-lab/songgot-l with Docker Model Runner:
```
docker model run hf.co/palette-lab/songgot-l:Q4_K_M
```

Lemonade

How to use palette-lab/songgot-l with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull palette-lab/songgot-l:Q4_K_M

Run and chat with the model

lemonade run user.songgot-l-Q4_K_M

List all available models

lemonade list

Atomic Chat

Hanish commited on 5 days ago

Commit

d858278

verified ·

1 Parent(s): 17f09a2

Model card: 12B tokens with instruction bucket, 2 epochs, v8 set

Browse files

Files changed (1) hide show

README.md +4 -4

README.md CHANGED Viewed

@@ -31,9 +31,9 @@ Kakao FunctionChat-Bench SingleCall (500 Korean items, 5 tool conditions), exact
 | model | params | exact | 4_random | 4_close | 8_random | 8_close | all | name only |
 |---|---|---|---|---|---|---|---|---|
-| Songgot-L (24B tokens, 2 epochs, v8 set) | 303M | 42.0 | 38.0 | 24.0 | 27.0 | 10.0 | 28.2 | 72.4 |
 | Songgot-M (2 epochs, v8 set) | 126M | 45.0 | 43.0 | 26.0 | 34.0 | 18.0 | 33.2 | 73.8 |
-| Songgot (6B tokens, 2 epochs, v8 set) | 50M | 44.0 | 39.0 | 30.0 | 35.0 | 17.0 | 33.0 | 73.6 |
 | Songgot-nano (1 epoch) | 39M | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
 | Needle 2 | 45M | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
 | FunctionGemma-270M | 270M | 3.0 | 5.0 | 1.0 | 1.0 | 1.0 | 2.2 | 36.2 |
@@ -52,8 +52,8 @@ Tokens per Hangul syllable on the same 100 queries: Songgot 0.90, Gemma 3 0.98,
 ![Pretraining loss](https://hanishkeloth.github.io/songgot/fig3_loss.png)
-## Status (2026-09-12 21:59)
-Weights in this repo are Songgot-L, 24B tokens, 2 epochs, v8 set: 24 layers, hidden 1024, about 303M parameters, trained from scratch on 8xH100 (Modal) on 24B tokens of FineWeb-2 Korean, Korean Wikipedia and fineweb-edu, post-trained on the v8 set, post-trained on the v2 tool-calling set. Call accuracy on FunctionChat-Bench SingleCall 28.2 percent (name only 72.4). GGUF exports (f16, Q8_0, Q4_K_M) are in this repo.
 ## Format
 ```
 <|system|>

 | model | params | exact | 4_random | 4_close | 8_random | 8_close | all | name only |
 |---|---|---|---|---|---|---|---|---|
+| Songgot-L (12B tokens with instruction bucket, 2 epochs, v8 set) | 303M | 40.0 | 36.0 | 21.0 | 30.0 | 20.0 | 29.4 | 75.0 |
 | Songgot-M (2 epochs, v8 set) | 126M | 45.0 | 43.0 | 26.0 | 34.0 | 18.0 | 33.2 | 73.8 |
+| Songgot (6B tokens, 3 epochs, v8 set) | 50M | 44.0 | 39.0 | 30.0 | 35.0 | 17.0 | 33.0 | 73.6 |
 | Songgot-nano (1 epoch) | 39M | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
 | Needle 2 | 45M | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
 | FunctionGemma-270M | 270M | 3.0 | 5.0 | 1.0 | 1.0 | 1.0 | 2.2 | 36.2 |
 ![Pretraining loss](https://hanishkeloth.github.io/songgot/fig3_loss.png)
+## Status (2026-09-13 09:33)
+Weights in this repo are Songgot-L, 12B tokens with instruction bucket, 2 epochs, v8 set: 24 layers, hidden 1024, about 303M parameters, trained from scratch on 8xH100 (Modal) on 12B tokens of Korean Wikipedia and fineweb-edu with 5 percent tool-calling rows in the mix, post-trained on the v8 set, post-trained on the v2 tool-calling set. Call accuracy on FunctionChat-Bench SingleCall 29.4 percent (name only 75.0). GGUF exports (f16, Q8_0, Q4_K_M) are in this repo.
 ## Format
 ```
 <|system|>