Instructions to use DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 with Transformers:

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

pipe = pipeline("text-generation", model="DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2")
model = AutoModelForCausalLM.from_pretrained("DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2

SGLang

How to use DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 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 "DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 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 DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 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 DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2",
    max_seq_length=2048,
)

Docker Model Runner
How to use DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2 with Docker Model Runner:
```
docker model run hf.co/DavidAU/ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

ERNIE-21B-A3B-Thinking-Gemini-3-Pro-High-Reasoning-V2

This is a uncensored, full deep thinking Ernie 21B-A3B (MOE, 64 experts) fine tune using Gemini 3 Pro High reasoning dataset via Unsloth via local hardware, Linux (for windows).

Note this model is mostly uncensored right from the "factory" so to speak.

Model excels in creative (brainstorming, creative prose) as well as general usage.

Reasoning is compact, but detailed (very detailed) and right to the "point" so to speak.

CRITICAL SETTINGS:

for creative suggest rep pen of 1.01 to 1.1
for general work rep pen of 1 (off), 1.05 or 1.1
Lower quants MAY LOOP in some cases.

Reasoning affects:

General model operation.
Output generation
Benchmarks.

Model Features:

128k context
Temp range .1 to 2.5.
Reasoning is temp stable.

You may want to visit Baidu's repo for this model for root/core benchmarks and settings.

https://huggingface.co/baidu/ERNIE-4.5-21B-A3B-Thinking

Enjoy the freedom!

BENCHMARKS:

arc_challenge,arc_easy,boolq,hellaswag,openbookqa,piqa,   winogrande

0.373        ,0.444   ,0.622,0.679    ,0.362     ,0.758  ,0.642

VS (regular model):

0.331       ,0.440   ,0.628,0.663    ,0.338     ,0.725   ,0.567

SPECIAL THANKS TO:

Team "TeichAI" for the excellent dataset.
Team "Unsloth" for making the training painless.
Team "Nightmedia" for Benchmarks and co-labing.

Settings: CHAT / ROLEPLAY and/or SMOOTHER operation of this model:

In "KoboldCpp" or "oobabooga/text-generation-webui" or "Silly Tavern" ;

Set the "Smoothing_factor" to 1.5

: in KoboldCpp -> Settings->Samplers->Advanced-> "Smooth_F"

: in text-generation-webui -> parameters -> lower right.

: In Silly Tavern this is called: "Smoothing"

NOTE: For "text-generation-webui"

-> if using GGUFs you need to use "llama_HF" (which involves downloading some config files from the SOURCE version of this model)

Source versions (and config files) of my models are here:

https://huggingface.co/collections/DavidAU/d-au-source-files-for-gguf-exl2-awq-gptq-hqq-etc-etc-66b55cb8ba25f914cbf210be

OTHER OPTIONS:

Increase rep pen to 1.1 to 1.15 (you don't need to do this if you use "smoothing_factor")
If the interface/program you are using to run AI MODELS supports "Quadratic Sampling" ("smoothing") just make the adjustment as noted.

Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers

This a "Class 1" model:

For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see:

[ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ]

You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here:

[ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ]