Instructions to use destr8803/GLM-5.2-224E-FP8-REAP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use destr8803/GLM-5.2-224E-FP8-REAP with Transformers:

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

pipe = pipeline("text-generation", model="destr8803/GLM-5.2-224E-FP8-REAP")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("destr8803/GLM-5.2-224E-FP8-REAP")
model = AutoModelForCausalLM.from_pretrained("destr8803/GLM-5.2-224E-FP8-REAP", 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 destr8803/GLM-5.2-224E-FP8-REAP with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "destr8803/GLM-5.2-224E-FP8-REAP"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "destr8803/GLM-5.2-224E-FP8-REAP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/destr8803/GLM-5.2-224E-FP8-REAP

SGLang

How to use destr8803/GLM-5.2-224E-FP8-REAP 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 "destr8803/GLM-5.2-224E-FP8-REAP" \
    --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": "destr8803/GLM-5.2-224E-FP8-REAP",
		"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 "destr8803/GLM-5.2-224E-FP8-REAP" \
        --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": "destr8803/GLM-5.2-224E-FP8-REAP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use destr8803/GLM-5.2-224E-FP8-REAP with Docker Model Runner:
```
docker model run hf.co/destr8803/GLM-5.2-224E-FP8-REAP
```

GLM-5.2-224E-FP8-REAP

This is an independently produced, expert-pruned derivative of zai-org/GLM-5.2-FP8. It is not an official Z.ai release.

The checkpoint reduces each sparse MoE layer from 256 routed experts to 224 while preserving 8 active experts per token. Experts were selected with REAP scores computed from actual router weights and unweighted expert-output norms. The weights retain the source model's block-scaled FP8 format.

Model Details

Property	Value
Base model	`zai-org/GLM-5.2-FP8`
Architecture	`GlmMoeDsaForCausalLM`
Routed experts	224 per sparse layer, reduced from 256
Active experts	8 per token
Hidden layers	78
Weight format	FP8, block size 128 x 128
Weight shards	141 Safetensors files
Weight payload	663,759,907,008 bytes (618.17 GiB)
Source payload	755,617,140,416 bytes
Payload reduction	12.2%
License	MIT, inherited from the base model

Pruning Method

REAP calibration used 12,228 samples and 15,425,969 tokens across general, code, reasoning, and agentic data. For every routed expert, the score was the conditional mean of:

actual_router_weight * L2(unweighted_expert_output)

The 32 lowest-scoring experts in each of the 75 sparse layers were removed. The derivative was built directly from the immutable 256-expert source; it was not produced by repeatedly pruning another derivative.

Keep-map SHA-256: 6845c6d68d39934a5866b12b1222cd29da214b1080485cf39d8a9986b7c203f6

Validation

Deep checkpoint validation: 104,037 of 104,037 tensors passed, with zero errors.
Runtime functional checks: 4 of 4 passed, covering arithmetic, exact instruction following, long-context retrieval, and structured tool calling.
Maximum context validated on the tested runtime: 131,072 tokens with BF16 KV cache.
Aggregate context matrix passed at 16 x 8K, 8 x 16K, 4 x 32K, 2 x 64K, and 1 x 126K.
Sequential decode: 20.8689 output tokens/s on the tested 8-HPU setup.
Concurrent decode: 263.5073 aggregate output tokens/s for 16 requests with 128 output tokens each.

Preliminary GPQA Check

A 10-question GPQA Diamond smoke sample scored 9/10 (90.0%) with no request errors or malformed answers. The published base-model score is 91.2%, but this small sample is not statistically sufficient for an exact quality comparison. It used the OpenAI simple-evals zero-shot chain-of-thought prompt with temperature=1.0 and top_p=0.95.

Deployment

The checkpoint was qualified with vLLM-Gaudi using tensor parallel 8, expert parallel 8, BF16 KV cache, and MTP disabled. A representative configuration is:

vllm serve destr8803/GLM-5.2-224E-FP8-REAP \
  --tensor-parallel-size 8 \
  --enable-expert-parallel \
  --kv-cache-dtype auto \
  --max-model-len 131072

Hardware capacity, graph configuration, and supported context length depend on the inference backend. The architecture advertises a larger nominal context, but this derivative has only been runtime-qualified through 131,072 tokens.

GLM-5.2 uses thinking by default. Keep the default maximum reasoning effort for benchmark reproduction; explicitly configure the serving framework if a lower reasoning effort or disabled thinking is desired.

Limitations

This is a structural MoE pruning derivative. Removing experts can affect quality unevenly across domains even when aggregate smoke tests pass.
The full official GLM-5.2 benchmark suite has not been reproduced.
The preliminary 10-question GPQA result must not be interpreted as a complete benchmark result.
Only the BF16 KV-cache runtime profile is quality-qualified. FP8 KV cache was not accepted for this release.
Deployment requires substantial storage and accelerator memory.

The base model's intended-use, safety, and language limitations continue to apply. Review the zai-org/GLM-5.2 model card before deployment.

License and Attribution

This derivative is distributed under the MIT license included in the repository. GLM-5.2 was created by Z.ai and Tsinghua University. Please cite the original technical report:

@misc{glm5team2026glm5vibecodingagentic,
  title={GLM-5: from Vibe Coding to Agentic Engineering},
  author={GLM-5-Team},
  year={2026},
  eprint={2602.15763},
  archivePrefix={arXiv},
  primaryClass={cs.LG},
  url={https://arxiv.org/abs/2602.15763}
}

Downloads last month: 19

Safetensors

Model size

662B params

Tensor type

BF16

F8_E4M3

F32

Model tree for destr8803/GLM-5.2-224E-FP8-REAP

Base model

zai-org/GLM-5.2-FP8

Quantized

(20)

this model

Paper for destr8803/GLM-5.2-224E-FP8-REAP

GLM-5: from Vibe Coding to Agentic Engineering

Paper • 2602.15763 • Published Feb 17 • 217