Instructions to use intervitens/internlm2-limarp-chat-20b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use intervitens/internlm2-limarp-chat-20b with Transformers:

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

pipe = pipeline("text-generation", model="intervitens/internlm2-limarp-chat-20b")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("intervitens/internlm2-limarp-chat-20b")
model = AutoModelForCausalLM.from_pretrained("intervitens/internlm2-limarp-chat-20b", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use intervitens/internlm2-limarp-chat-20b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "intervitens/internlm2-limarp-chat-20b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "intervitens/internlm2-limarp-chat-20b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/intervitens/internlm2-limarp-chat-20b

SGLang

How to use intervitens/internlm2-limarp-chat-20b 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 "intervitens/internlm2-limarp-chat-20b" \
    --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": "intervitens/internlm2-limarp-chat-20b",
		"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 "intervitens/internlm2-limarp-chat-20b" \
        --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": "intervitens/internlm2-limarp-chat-20b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use intervitens/internlm2-limarp-chat-20b with Docker Model Runner:
```
docker model run hf.co/intervitens/internlm2-limarp-chat-20b
```

Experimental model, LimaRP LoRA trained on top of internlm2-base-20b with 8192 context length and merged with internlm2-chat-20b.

Prompt format is ChatML.

internlm2-06limarp-1chat-TASK_ARITHM-20b-v0.03

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the task arithmetic merge method using intervitens/internlm2-base-20b-llama as a base.

Models Merged

The following models were included in the merge:

./internlm2-limarp-20b-v0.03
./internlm2-chat-20b-llama

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: ./internlm2-chat-20b-llama
    parameters:
      weight: 1.0
  - model: ./internlm2-limarp-20b-v0.03
    parameters:
      weight: 0.6
merge_method: task_arithmetic
base_model: ./internlm2-base-20b-llama
parameters:
  #normalize: false
  #int8_mask: true
dtype: bfloat16