Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
Paper • 2311.03099 • Published • 33
How to use appvoid/dot-v1.4 with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="appvoid/dot-v1.4") # Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("appvoid/dot-v1.4")
model = AutoModelForCausalLM.from_pretrained("appvoid/dot-v1.4")How to use appvoid/dot-v1.4 with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "appvoid/dot-v1.4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "appvoid/dot-v1.4",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/appvoid/dot-v1.4
How to use appvoid/dot-v1.4 with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "appvoid/dot-v1.4" \
--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": "appvoid/dot-v1.4",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'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 "appvoid/dot-v1.4" \
--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": "appvoid/dot-v1.4",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use appvoid/dot-v1.4 with Docker Model Runner:
docker model run hf.co/appvoid/dot-v1.4
This is a merge of pre-trained language models created using mergekit.
This model was merged using the DARE TIES merge method using appvoid/palmer-003 as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: vihangd/DopeyTinyLlama-1.1B-v1
parameters:
density: 0.25
weight: 0.25
- model: raidhon/coven_tiny_1.1b_32k_orpo_alpha
parameters:
density: 0.25
weight: 0.25
- model: l3utterfly/tinyllama-1.1b-layla-v4
parameters:
density: 0.25
weight: 0.25
- model: ShieldX/manovyadh-1.1B-v1-chat
parameters:
density: 0.25
weight: 0.25
merge_method: dare_ties
base_model: appvoid/palmer-003
parameters:
normalize: true
int8_mask: true
dtype: float16