Instructions to use Entropicengine/Trifecta-L3-8b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Entropicengine/Trifecta-L3-8b with Transformers:

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

pipe = pipeline("text-generation", model="Entropicengine/Trifecta-L3-8b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Entropicengine/Trifecta-L3-8b")
model = AutoModelForCausalLM.from_pretrained("Entropicengine/Trifecta-L3-8b")
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]:]))

Inference
Local Apps Settings

vLLM

How to use Entropicengine/Trifecta-L3-8b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Entropicengine/Trifecta-L3-8b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Entropicengine/Trifecta-L3-8b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Entropicengine/Trifecta-L3-8b

SGLang

How to use Entropicengine/Trifecta-L3-8b 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 "Entropicengine/Trifecta-L3-8b" \
    --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": "Entropicengine/Trifecta-L3-8b",
		"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 "Entropicengine/Trifecta-L3-8b" \
        --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": "Entropicengine/Trifecta-L3-8b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Entropicengine/Trifecta-L3-8b with Docker Model Runner:
```
docker model run hf.co/Entropicengine/Trifecta-L3-8b
```

Decent model

by yano2mch - opened Aug 19, 2025

Discussion

yano2mch

Aug 19, 2025

•

edited Aug 19, 2025

Noted from a Quant request located here: https://huggingface.co/mradermacher/model_requests/discussions/1289#68a3427143e57453205ad661

After searching and testing many 8B models I finally found what I was looking for. An absolute all-rounder

So far i'm playing around with it and... i think @hollywood2011 's right, it's overall a decent all-round 8B model that does good with RPing. I've had it explain and do code generation (which it loves python for some reason), RPing, song generation among some of the outputs.

Best of all because it's small it's incredibly fast and i can fit almost all of it (i1-Q6) in my 8Gb video card running it purely on GPU i get 10t/s.

I'll have to test the Q3 variant to see if it's reliable enough, if so I could put it on a DVD (with oobabooga's webui).

yano2mch

Aug 23, 2025

Curious, this model is pretty good at making lyrics for music.

Other models that are better for RP feel they fall quite short.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment