Instructions to use meta-llama/Llama-3.1-8B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use meta-llama/Llama-3.1-8B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="meta-llama/Llama-3.1-8B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
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
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use meta-llama/Llama-3.1-8B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "meta-llama/Llama-3.1-8B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "meta-llama/Llama-3.1-8B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/meta-llama/Llama-3.1-8B-Instruct

SGLang

How to use meta-llama/Llama-3.1-8B-Instruct 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 "meta-llama/Llama-3.1-8B-Instruct" \
    --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": "meta-llama/Llama-3.1-8B-Instruct",
		"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 "meta-llama/Llama-3.1-8B-Instruct" \
        --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": "meta-llama/Llama-3.1-8B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use meta-llama/Llama-3.1-8B-Instruct with Docker Model Runner:
```
docker model run hf.co/meta-llama/Llama-3.1-8B-Instruct
```

Cross-architecture RYS sweep — Llama-3.1-8B-Instruct (richer multi-circuit than Qwen2.5-7B at same scale; 15 of 66 boosters)

#370

by john-broadway - opened 17 days ago

Discussion

john-broadway

17 days ago

Sharing a cross-architecture RYS (layer-duplication, "Repeat Your Self") sweep that includes Llama-3.1-8B-Instruct alongside 20 other model variants spanning 10 architecture families.

Sweep result for this model (32 layers, Q4_K_M, baseline reasoning 82.35%):

Configuration	Math Δ	EQ Δ	Reasoning Δ
Best: (18,22) block-4	+6.35	−5.59	+11.76

Peak reasoning Δ: +17.65%, with 15 of 66 configurations boosting reasoning >5%. At a comparable scale and matched baseline, Qwen2.5-7B-Instruct shows only 5 boosting configurations — Llama-3.1 appears to carry a richer multi-circuit structure that absorbs RYS-style duplication without breaking.

The cross-architecture finding (Pearson r = −0.726 across 21 variants, 10 families): weak baselines lift more, in their weakest dimension. Llama-3.1-8B sits at the high-baseline end of the curve, where lifts are necessarily modest but the number of boosting configurations carries the multi-circuit signal.

Full sweep data + analysis: https://huggingface.co/datasets/john-broadway/rys-sovereign-collection-v2
Evaluation card for Llama-3.1-8B-Instruct: https://huggingface.co/john-broadway/Llama-3.1-8B-RYS-eval

Method: original RYS post by David Ng; sweep toolkit by alainnothere. Train-free — no weight changes, no merging.

— John Broadway, with collaboration from Claude (Opus 4.6 in April 2026 sweep generation; Opus 4.7 in May 2026 cross-architecture analysis).

john-broadway

17 days ago

Update (2026-05-13 PM): The eval-only john-broadway/Llama-3.1-8B-RYS-eval repo linked in the original post has been consolidated. The same sweep results + mechanism writeup are now in the deployable weights repo: john-broadway/Llama-3.1-8B-RYS-18-22-GGUF — RYS-applied Q4_K_M GGUF, ready for llama-server. No new content, just one repo per model instead of two.

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