Instructions to use JSpergel/test_tiny_mixtral_only_router with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use JSpergel/test_tiny_mixtral_only_router with Transformers:

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

pipe = pipeline("text-generation", model="JSpergel/test_tiny_mixtral_only_router")

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

tokenizer = AutoTokenizer.from_pretrained("JSpergel/test_tiny_mixtral_only_router")
model = AutoModelForCausalLM.from_pretrained("JSpergel/test_tiny_mixtral_only_router", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use JSpergel/test_tiny_mixtral_only_router with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/JSpergel/test_tiny_mixtral_only_router

SGLang

How to use JSpergel/test_tiny_mixtral_only_router 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 "JSpergel/test_tiny_mixtral_only_router" \
    --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": "JSpergel/test_tiny_mixtral_only_router",
		"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 "JSpergel/test_tiny_mixtral_only_router" \
        --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": "JSpergel/test_tiny_mixtral_only_router",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use JSpergel/test_tiny_mixtral_only_router with Docker Model Runner:
```
docker model run hf.co/JSpergel/test_tiny_mixtral_only_router
```

test_tiny_mixtral_only_router / README.md

JSpergel

Update README.md

adb4054 verified over 2 years ago

preview code

Raw History Blame Contribute Delete

1.6 kB

metadata

license: apache-2.0
tags:
  - moe
  - frankenmoe
  - merge
  - mergekit
  - lazymergekit
  - openaccess-ai-collective/tiny-mistral
base_model:
  - openaccess-ai-collective/tiny-mistral
  - openaccess-ai-collective/tiny-mistral
  - openaccess-ai-collective/tiny-mistral
  - openaccess-ai-collective/tiny-mistral

test_tiny_mixtral_only_router

test_tiny_mixtral_only_router is a Mixure of Experts (MoE) made with the following models using a modified version of mergekit.

🧩 Configuration

base_model: openaccess-ai-collective/tiny-mistral
gate_mode: hidden
dtype: bfloat16
experts:
  - source_model: openaccess-ai-collective/tiny-mistral
    positive_prompts:
      - "math"
    # You can add negative_prompts if needed
  - source_model: openaccess-ai-collective/tiny-mistral

    positive_prompts:
      - "science"
  - source_model: openaccess-ai-collective/tiny-mistral
    positive_prompts:
      - "writing"
    # You can add negative_prompts if needed
  - source_model: openaccess-ai-collective/tiny-mistral
    positive_prompts:
      - "general"

This is a test version of arcee-ai's hidden state model. It is a router for a frankenMoE instead of the entire MoE itself