Text Generation
Transformers
Safetensors
GGUF
Korean
English
llama
3b
korean
from-scratch
orpo
instruction-tuned
preference-aligned
fp8
b200
Eval Results (legacy)
text-generation-inference
Instructions to use pathcosmos/frankenstallm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use pathcosmos/frankenstallm with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="pathcosmos/frankenstallm")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("pathcosmos/frankenstallm") model = AutoModelForCausalLM.from_pretrained("pathcosmos/frankenstallm") - llama-cpp-python
How to use pathcosmos/frankenstallm with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="pathcosmos/frankenstallm", filename="gguf/frankenstallm-3b-Q4_K_M.gguf", )
output = llm( "Once upon a time,", max_tokens=512, echo=True ) print(output)
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- llama.cpp
How to use pathcosmos/frankenstallm with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf pathcosmos/frankenstallm:Q4_K_M # Run inference directly in the terminal: llama-cli -hf pathcosmos/frankenstallm:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf pathcosmos/frankenstallm:Q4_K_M # Run inference directly in the terminal: llama-cli -hf pathcosmos/frankenstallm: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 pathcosmos/frankenstallm:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf pathcosmos/frankenstallm: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 pathcosmos/frankenstallm:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf pathcosmos/frankenstallm:Q4_K_M
Use Docker
docker model run hf.co/pathcosmos/frankenstallm:Q4_K_M
- LM Studio
- Jan
- vLLM
How to use pathcosmos/frankenstallm with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "pathcosmos/frankenstallm" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "pathcosmos/frankenstallm", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/pathcosmos/frankenstallm:Q4_K_M
- SGLang
How to use pathcosmos/frankenstallm 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 "pathcosmos/frankenstallm" \ --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": "pathcosmos/frankenstallm", "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 "pathcosmos/frankenstallm" \ --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": "pathcosmos/frankenstallm", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Ollama
How to use pathcosmos/frankenstallm with Ollama:
ollama run hf.co/pathcosmos/frankenstallm:Q4_K_M
- Unsloth Studio new
How to use pathcosmos/frankenstallm 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 pathcosmos/frankenstallm 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 pathcosmos/frankenstallm to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for pathcosmos/frankenstallm to start chatting
- Docker Model Runner
How to use pathcosmos/frankenstallm with Docker Model Runner:
docker model run hf.co/pathcosmos/frankenstallm:Q4_K_M
- Lemonade
How to use pathcosmos/frankenstallm with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull pathcosmos/frankenstallm:Q4_K_M
Run and chat with the model
lemonade run user.frankenstallm-Q4_K_M
List all available models
lemonade list
| # ============================================================================= | |
| # launch_korean_3b.sh โ 8-GPU FP8 pretraining launcher for 3B Korean LLM | |
| # | |
| # Usage: | |
| # bash scripts/launch_korean_3b.sh # full run (~60B tokens) | |
| # bash scripts/launch_korean_3b.sh --max_steps 50 # quick benchmark | |
| # bash scripts/launch_korean_3b.sh --resume checkpoints/korean_3b_fp8_run1/checkpoint-XXXXX | |
| # | |
| # Effective batch size: 8 (local) ร 8 GPU ร 4 (grad_accum) ร 4096 (seq_len) | |
| # = 1,048,576 tokens / step | |
| # ============================================================================= | |
| set -euo pipefail | |
| RUN_NAME="${RUN_NAME:-korean_3b_fp8_run1}" | |
| CONFIG="${CONFIG:-configs/3b_pretrain.yaml}" | |
| CKPT_DIR="checkpoints/${RUN_NAME}" | |
| LOG_FILE="${CKPT_DIR}/train.log" | |
| NPROC=8 | |
| MASTER_PORT="${MASTER_PORT:-29502}" | |
| MAX_STEPS=57000 | |
| BATCH_SIZE=4 | |
| GRAD_ACCUM=8 | |
| LR=1.5e-4 | |
| WARMUP_STEPS=2000 | |
| SEED=42 | |
| EXTRA_ARGS="$@" | |
| # ---- B200 / NVSwitch NCCL tuning ------------------------------------------- | |
| export NCCL_IB_DISABLE=1 | |
| export NCCL_ALGO=Ring | |
| export NCCL_PROTO=Simple | |
| export NCCL_MIN_NCHANNELS=16 | |
| export NCCL_MAX_NCHANNELS=16 | |
| export NCCL_BUFFSIZE=67108864 | |
| export OMP_NUM_THREADS=4 | |
| export MKL_NUM_THREADS=4 | |
| # cd FIRST โ ์ดํ ์๋๊ฒฝ๋ก ์ฒดํฌ๊ฐ ํ๋ก์ ํธ ๋ฃจํธ ๊ธฐ์ค์ผ๋ก ๋์ | |
| cd "$(dirname "$0")/.." | |
| # TRAIN_DATA fallback: cd ์ดํ์ ์๋๊ฒฝ๋ก ์ฒดํฌ | |
| if [[ -f "data/merged_3b_train.bin" ]]; then | |
| TRAIN_DATA="${TRAIN_DATA:-data/merged_3b_train.bin}" | |
| echo "Using merged training data: data/merged_3b_train.bin" | |
| elif [[ -f "data/korean_train.bin" ]]; then | |
| TRAIN_DATA="${TRAIN_DATA:-data/korean_train.bin}" | |
| echo "Using fallback training data: data/korean_train.bin" | |
| else | |
| echo "ERROR: No training data found (data/merged_3b_train.bin or data/korean_train.bin)" | |
| exit 1 | |
| fi | |
| # VAL_DATA fallback: cd ์ดํ์ ์๋๊ฒฝ๋ก ์ฒดํฌ | |
| VAL_DATA="${VAL_DATA:-data/merged_3b_val.bin}" | |
| if [[ ! -f "${VAL_DATA}" ]]; then | |
| VAL_DATA="data/korean_val.bin" | |
| fi | |
| if [[ ! -f "${TRAIN_DATA}" ]]; then | |
| echo "ERROR: Training data not found: ${TRAIN_DATA}" | |
| exit 1 | |
| fi | |
| if [[ ! -f "${VAL_DATA}" ]]; then | |
| echo "ERROR: Validation data not found: ${VAL_DATA}" | |
| exit 1 | |
| fi | |
| mkdir -p "${CKPT_DIR}" | |
| echo "==================================================================" | |
| echo " Run name : ${RUN_NAME}" | |
| echo " Config : ${CONFIG}" | |
| echo " Train data : ${TRAIN_DATA}" | |
| echo " CKPT dir : ${CKPT_DIR}" | |
| echo " Max steps : ${MAX_STEPS}" | |
| echo " LR : ${LR}" | |
| echo " Batch size : ${BATCH_SIZE} (local) ร ${NPROC} GPU ร ${GRAD_ACCUM} grad_accum" | |
| echo " Started : $(date)" | |
| echo "==================================================================" | |
| export PYTHONWARNINGS="ignore::UserWarning:torch.library" | |
| torchrun \ | |
| --nproc_per_node=${NPROC} \ | |
| --master_port=${MASTER_PORT} \ | |
| train/pretrain.py \ | |
| --config "${CONFIG}" \ | |
| --train_data "${TRAIN_DATA}" \ | |
| --val_data "${VAL_DATA}" \ | |
| --checkpoint_dir "${CKPT_DIR}" \ | |
| --log_file "${LOG_FILE}" \ | |
| --max_steps ${MAX_STEPS} \ | |
| --batch_size ${BATCH_SIZE} \ | |
| --lr ${LR} \ | |
| --grad_accum ${GRAD_ACCUM} \ | |
| --warmup_steps ${WARMUP_STEPS} \ | |
| --seed ${SEED} \ | |
| ${EXTRA_ARGS} \ | |
| 2>&1 | grep -v "UserWarning" \ | |
| | grep -v "Warning only once" \ | |
| | grep -v "Overriding a previously" \ | |
| | grep -v "dispatch key:" \ | |
| | grep -v "previous kernel:" \ | |
| | grep -v "new kernel:" \ | |
| | grep -v "operator: flash_attn" \ | |
| | grep -v "registered at /usr/local" \ | |
| | grep -v "self.m.impl" \ | |
| | tee -a "${LOG_FILE}" | |
| echo "==================================================================" | |
| echo " Done : $(date)" | |
| echo "==================================================================" | |