Instructions to use AvoCahDoe/llama-3-8b-rlmpq with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AvoCahDoe/llama-3-8b-rlmpq with Transformers:

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

pipe = pipeline("text-generation", model="AvoCahDoe/llama-3-8b-rlmpq")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AvoCahDoe/llama-3-8b-rlmpq")
model = AutoModelForCausalLM.from_pretrained("AvoCahDoe/llama-3-8b-rlmpq", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AvoCahDoe/llama-3-8b-rlmpq with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AvoCahDoe/llama-3-8b-rlmpq"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AvoCahDoe/llama-3-8b-rlmpq",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/AvoCahDoe/llama-3-8b-rlmpq

SGLang

How to use AvoCahDoe/llama-3-8b-rlmpq 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 "AvoCahDoe/llama-3-8b-rlmpq" \
    --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": "AvoCahDoe/llama-3-8b-rlmpq",
		"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 "AvoCahDoe/llama-3-8b-rlmpq" \
        --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": "AvoCahDoe/llama-3-8b-rlmpq",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use AvoCahDoe/llama-3-8b-rlmpq with Docker Model Runner:
```
docker model run hf.co/AvoCahDoe/llama-3-8b-rlmpq
```

llama-3-8b-rlmpq / Extreme_Survival /README.md

AvoCahDoe

RL-MPQ Extreme_Survival — 2026-06-11T15:42:55.778240

a6aa222 verified about 2 months ago

preview code

Raw

History Blame Contribute Delete

1.24 kB

	---
	license: llama3
	base_model: meta-llama/Meta-Llama-3-8B
	tags:
	- rl-mpq
	- mixed-precision
	- quantization
	library_name: transformers
	---

	# LLAMA-3-8B — Extreme Survival

	Subfolder of [`AvoCahDoe/llama-3-8b-rlmpq`](https://huggingface.co/AvoCahDoe/llama-3-8b-rlmpq) (all RL-MPQ scenarios for this model).

	Fake-quantized weights for Extreme_Survival applied to
	[meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) (per-layer asymmetric group-wise quant, group_size=128).

	## Load

	```python
	from transformers import AutoModelForCausalLM, AutoTokenizer

	repo = "AvoCahDoe/llama-3-8b-rlmpq"
	model = AutoModelForCausalLM.from_pretrained(repo, subfolder="Extreme_Survival", torch_dtype="float16")
	tokenizer = AutoTokenizer.from_pretrained(repo, subfolder="Extreme_Survival")
	```

	## Metrics

	\| Metric \| Value \|
	\|--------\|-------\|
	\| Scenario \| Extreme_Survival \|
	\| Average bits per weight \| 2.875 \|
	\| Compression ratio (vs FP16) \| 5.5652x \|
	\| WikiText-2 perplexity \| 32.393 \|
	\| Layers \| 32 \|
	\| Bit distribution \| {'2': 4, '3': 28} \|

	## Policy (bits per layer)

	```
	[3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2]
	```

	Generated: 2026-06-11T15:42:55.694321