Instructions to use marksverdhei/GLM-4.7-Flash-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use marksverdhei/GLM-4.7-Flash-FP8 with Transformers:

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

pipe = pipeline("text-generation", model="marksverdhei/GLM-4.7-Flash-FP8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("marksverdhei/GLM-4.7-Flash-FP8")
model = AutoModelForCausalLM.from_pretrained("marksverdhei/GLM-4.7-Flash-FP8", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use marksverdhei/GLM-4.7-Flash-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "marksverdhei/GLM-4.7-Flash-FP8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "marksverdhei/GLM-4.7-Flash-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/marksverdhei/GLM-4.7-Flash-FP8

SGLang

How to use marksverdhei/GLM-4.7-Flash-FP8 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 "marksverdhei/GLM-4.7-Flash-FP8" \
    --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": "marksverdhei/GLM-4.7-Flash-FP8",
		"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 "marksverdhei/GLM-4.7-Flash-FP8" \
        --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": "marksverdhei/GLM-4.7-Flash-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use marksverdhei/GLM-4.7-Flash-FP8 with Docker Model Runner:
```
docker model run hf.co/marksverdhei/GLM-4.7-Flash-FP8
```

GLM-4.7-Flash-FP8 / README.md

marksverdhei

Update README.md

1f6f51d verified 8 months ago

preview code

Raw

History Blame Contribute Delete

1.5 kB

metadata

license: mit
base_model: zai-org/GLM-4.7-Flash
tags:
  - fp8
  - quantized
  - glm4
  - moe
library_name: transformers

GLM-4.7-Flash FP8

FP8 quantized version of zai-org/GLM-4.7-Flash.

Also, see Unsloth's new unsloth/GLM-4.7-Flash-FP8-Dynamic

Quantization Details

Method: FP8 E4M3 per-tensor quantization with embedded scales
Original size: ~62GB (BF16)
Quantized size: ~30GB (FP8)
Preserved in BF16: lm_head, embed_tokens, layernorms, router weights

Performance

Tested on 2x RTX 3090 (24GB each) with vLLM 0.13.0:

Setting	Value
Tensor Parallel	2
Context Length	8192
VRAM per GPU	14.7 GB
Throughput	19.4 tokens/sec

Note: RTX 3090 lacks native FP8 support, so vLLM uses the Marlin kernel for weight-only FP8 decompression. GPUs with native FP8 (RTX 40xx, Ada Lovelace+) will achieve higher throughput.

Usage with vLLM

Requires vLLM 0.13.0+ and transformers 5.0+ for glm4_moe_lite architecture support.

from vllm import LLM, SamplingParams

llm = LLM(
    model="marksverdhei/GLM-4.7-Flash-fp8",
    tensor_parallel_size=2,
    max_model_len=8192,
    enforce_eager=True,  # Optional: disable CUDA graphs to save VRAM
)

outputs = llm.generate(["Hello, world!"], SamplingParams(max_tokens=100))
print(outputs[0].outputs[0].text)

License

MIT (same as base model)