SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX

Instructions to use Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX with Transformers:

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

pipe = pipeline("text-generation", model="Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX")
model = AutoModelForCausalLM.from_pretrained("Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX

SGLang

How to use Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX 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 "Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX" \
    --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": "Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX",
		"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 "Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX" \
        --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": "Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX with Docker Model Runner:
```
docker model run hf.co/Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX
```

Jiunsong commited on 14 days ago

Commit

8ed8bde

verified ·

1 Parent(s): c1641d9

Add sealed execution bundle manifest

Browse files

Files changed (1) hide show

repro/bundle_manifest.json +2 -2

repro/bundle_manifest.json CHANGED Viewed

@@ -114,9 +114,9 @@
       "sha256": "749eaa801b2672bd2345ba71f25e4fac5d60d070c72df90286813079d01b6cc3"
     },
     {
-      "bytes": 15984,
       "path": "scripts/check_superdeepseek_v4_release_gate.py",
-      "sha256": "1494a4bbda54bca66c34478937bc573db690bc1348bac00a0c4d0922e1711928"
     },
     {
       "bytes": 3711,

       "sha256": "749eaa801b2672bd2345ba71f25e4fac5d60d070c72df90286813079d01b6cc3"
     },
     {
+      "bytes": 16179,
       "path": "scripts/check_superdeepseek_v4_release_gate.py",
+      "sha256": "de8cb2962a06a0515eab600eafd94486ab4acd24f4856cb4f1361f7fc322f934"
     },
     {
       "bytes": 3711,