Apertus-70B-Instruct-2509-FP8-dynamic

Apertus-70B-Instruct-2509-FP8-dynamic

Model Overview

• Model Architecture: ApertusForCausalLM
  • Input: Text
  • Output: Text
• Model Optimizations:
  • Weight quantization: FP8
  • Activation quantization: FP8
• Release Date: 9/18/2025
• Version: 1.0
• Model Developers: Red Hat

Quantized version of swiss-ai/Apertus-70B-2509.

Model Optimizations

This model was obtained by quantizing the weights and activations of swiss-ai/Apertus-70B-2509 to FP8 data type. This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%. Only the weights and activations of the linear operators within transformers blocks are quantized.

Deployment

Use with vLLM

1. Initialize vLLM server:

vllm serve RedHatAI/Apertus-70B-Instruct-2509-FP8-dynamic --tensor_parallel_size 2

1. Send requests to the server:

from openai import OpenAI

# Modify OpenAI's API key and API base to use vLLM's API server.
openai_api_key = "EMPTY"
openai_api_base = "http://<your-server-host>:8000/v1"

client = OpenAI(
    api_key=openai_api_key,
    base_url=openai_api_base,
)

model = "RedHatAI/Apertus-70B-Instruct-2509-FP8-dynamic"

messages = [
    {"role": "user", "content": "Give me a short introduction to large language model."},
]

outputs = client.chat.completions.create(
    model=model,
    messages=messages,
)

generated_text = outputs.choices[0].message.content
print(generated_text)

Creation

This model was created with llm-compressor by running the code snippet below.

<details> <summary>Model Creation Code</summary>

from llmcompressor.modifiers.quantization import QuantizationModifier
from llmcompressor.transformers import oneshot
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model
model_stub = "swiss-ai/Apertus-70B-Instruct-2509"
model_name = model_stub.split("/")[-1]

model = AutoModelForCausalLM.from_pretrained(model_stub, dtype="auto")

tokenizer = AutoTokenizer.from_pretrained(model_stub)

# Configure the quantization algorithm and scheme
recipe = QuantizationModifier(
    ignore=["lm_head"],
    targets="Linear",
    scheme="FP8_dynamic",
)

# Apply quantization
oneshot(
    model=model,
    recipe=recipe,
)

# Save to disk in compressed-tensors format
save_path = model_name + "-FP8-dynamic"
model.save_pretrained(save_path)
tokenizer.save_pretrained(save_path)
print(f"Model and tokenizer saved to: {save_path}")

</details>

Evaluation

The model was evaluated on OpenLLM Leaderboard V1, using the following command:

<details> <summary>Evaluation Commands</summary>

OpenLLM Leaderboard V1:

lm_eval \
  --model vllm \
  --model_args pretrained="RedHatAI/Apertus-70B-Instruct-2509-FP8-dynamic",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=1,gpu_memory_utilization=0.6,enable_chunked_prefill=True \
  --tasks openllm \
  --write_out \
  --batch_size auto \
  --output_path output_dir \
  --show_config

</details>

Accuracy

<table> <thead> <tr> <th>Category</th> <th>Metric</th> <th>swiss-ai/Apertus-70B-Instruct-2509</th> <th>RedHatAI/Apertus-70B-Instruct-2509-FP8-dynamic</th> <th>Recovery (%)</th> </tr> </thead> <tbody> <!-- OpenLLM Leaderboard V1 --> <tr> <td rowspan="7"><b>OpenLLM V1</b></td> <td>ARC-Challenge (Acc-Norm, 25-shot)</td> <td>70.82</td> <td>70.56</td> <td>99.6</td> </tr> <tr> <td>GSM8K (Strict-Match, 5-shot)</td> <td>73.69</td> <td>73.09</td> <td>99.2</td> </tr> <tr> <td>HellaSwag (Acc-Norm, 10-shot)</td> <td>86.23</td> <td>85.89</td> <td>99.61</td> </tr> <tr> <td>MMLU (Acc, 5-shot)</td> <td>69.21</td> <td>68.25</td> <td>98.6</td> </tr> <tr> <td>TruthfulQA (MC2, 0-shot)</td> <td>60.31</td> <td>60.05</td> <td>99.6</td> </tr> <tr> <td>Winogrande (Acc, 5-shot)</td> <td>80.74</td> <td>80.74</td> <td>100.0</td> </tr> <tr> <td><b>Average Score</b></td> <td><b>73.50</b></td> <td><b>73.10</b></td> <td><b>99.5</b></td> </tr> </tbody> </table>