Instructions to use RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV with Transformers:

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

pipe = pipeline("text-generation", model="RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV")
model = AutoModelForCausalLM.from_pretrained("RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV", device_map="auto")
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
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV

SGLang

How to use RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV 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 "RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV" \
    --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": "RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV",
		"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 "RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV" \
        --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": "RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV with Docker Model Runner:
```
docker model run hf.co/RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV
```

Meta-Llama-3-8B-Instruct-FP8-KV

Model Overview

Meta-Llama-3-8B-Instruct quantized to FP8 weights and activations using per-tensor quantization, ready for inference with vLLM >= 0.5.0. This model checkpoint also includes per-tensor scales for FP8 quantized KV Cache, accessed through the --kv-cache-dtype fp8 argument in vLLM.

from vllm import LLM
model = LLM(model="neuralmagic/Meta-Llama-3-8B-Instruct-FP8-KV", kv_cache_dtype="fp8")
result = model.generate("Hello, my name is")

Usage and Creation

Produced using AutoFP8 with calibration samples from ultrachat.

from datasets import load_dataset
from transformers import AutoTokenizer

from auto_fp8 import AutoFP8ForCausalLM, BaseQuantizeConfig

pretrained_model_dir = "meta-llama/Meta-Llama-3-8B-Instruct"
quantized_model_dir = "Meta-Llama-3-8B-Instruct-FP8-KV"

tokenizer = AutoTokenizer.from_pretrained(pretrained_model_dir, use_fast=True)
tokenizer.pad_token = tokenizer.eos_token

ds = load_dataset("mgoin/ultrachat_2k", split="train_sft")
examples = [tokenizer.apply_chat_template(batch["messages"], tokenize=False) for batch in ds]
examples = tokenizer(examples, padding=True, truncation=True, return_tensors="pt").to("cuda")

quantize_config = BaseQuantizeConfig(
    quant_method="fp8",
    activation_scheme="static",
    ignore_patterns=["re:.*lm_head"],
    kv_cache_quant_targets=("k_proj", "v_proj"),
)

model = AutoFP8ForCausalLM.from_pretrained(pretrained_model_dir, quantize_config)
model.quantize(examples)
model.save_quantized(quantized_model_dir)

Evaluation

Open LLM Leaderboard evaluation scores

	Meta-Llama-3-8B-Instruct	Meta-Llama-3-8B-Instruct-FP8	Meta-Llama-3-8B-Instruct-FP8-KV (this model)
gsm8k 5-shot	75.44	74.37	74.98

Downloads last month: 23,315

Safetensors

Model size

8B params

Tensor type

BF16

F8_E4M3

Model tree for RedHatAI/Meta-Llama-3-8B-Instruct-FP8-KV

Base model

meta-llama/Meta-Llama-3-8B-Instruct

Quantized

(277)

this model