Instructions to use ukkathva/TelecomGPT-R1-27B-FP8-Dynamic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ukkathva/TelecomGPT-R1-27B-FP8-Dynamic with Transformers:

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

pipe = pipeline("text-generation", model="ukkathva/TelecomGPT-R1-27B-FP8-Dynamic")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("ukkathva/TelecomGPT-R1-27B-FP8-Dynamic")
model = AutoModelForMultimodalLM.from_pretrained("ukkathva/TelecomGPT-R1-27B-FP8-Dynamic", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ukkathva/TelecomGPT-R1-27B-FP8-Dynamic with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ukkathva/TelecomGPT-R1-27B-FP8-Dynamic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ukkathva/TelecomGPT-R1-27B-FP8-Dynamic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ukkathva/TelecomGPT-R1-27B-FP8-Dynamic

SGLang

How to use ukkathva/TelecomGPT-R1-27B-FP8-Dynamic 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 "ukkathva/TelecomGPT-R1-27B-FP8-Dynamic" \
    --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": "ukkathva/TelecomGPT-R1-27B-FP8-Dynamic",
		"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 "ukkathva/TelecomGPT-R1-27B-FP8-Dynamic" \
        --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": "ukkathva/TelecomGPT-R1-27B-FP8-Dynamic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ukkathva/TelecomGPT-R1-27B-FP8-Dynamic with Docker Model Runner:
```
docker model run hf.co/ukkathva/TelecomGPT-R1-27B-FP8-Dynamic
```

TelecomGPT-R1-27B-FP8-Dynamic

File size: 906 Bytes

8d16bb8

---
language:
  - en
library_name: transformers
pipeline_tag: text-generation
license: apache-2.0
base_model: KU-DFI/TelecomGPT-R1
base_model_relation: quantized
tags:
  - fp8
  - compressed-tensors
  - vllm
  - telecom
  - reasoning
---

# TelecomGPT-R1-27B-FP8-Dynamic

FP8 Dynamic quantization of `KU-DFI/TelecomGPT-R1`.

## Quantization

- Scheme: `FP8_DYNAMIC`
- Serialization: `compressed-tensors`
- Target modules: `Linear`
- `lm_head`: unquantized
- Calibration dataset: none
- Source precision: BF16

## Intended use

Telecom reasoning, alarm analysis, root-cause analysis, protocol reasoning,
and evaluation against operator-specific incident datasets.

## A100 note

NVIDIA A100 is an Ampere GPU and does not provide native Hopper-style FP8
Tensor Core execution. The FP8 checkpoint still reduces model-weight memory,
and vLLM can use its supported Ampere execution path when loading the model.