Instructions to use wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1 with Transformers:

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

pipe = pipeline("image-text-to-text", model="wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1")
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("wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1")
model = AutoModelForMultimodalLM.from_pretrained("wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1", 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 wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1

SGLang

How to use wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1 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 "wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1" \
    --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": "wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1" \
        --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": "wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1 with Docker Model Runner:
```
docker model run hf.co/wrldsuksgo2mars/GLM-5.3-Flash-EXL3-K3.25-v1
```

wrldsuksgo2mars commited on Sep 5

Commit

59484d5

verified ·

1 Parent(s): 701cd74

docs: add final v0.7.0 runtime qualification

Browse files

Files changed (1) hide show

README.md +18 -8

README.md CHANGED Viewed

@@ -43,14 +43,24 @@ recovery contract covers any expert below the 1,024-route floor. Allocation
 provenance and the validation report ship with the model; the full error ledger
 is retained separately as internal quantization evidence.
-The first weight publication is gated on the exact checkpoint, allocation,
-native-byte, manifest, and configuration-parser audits above. The matching
-two-GPU B12x/vLLM release is separately gated on loading this materialized
-checkpoint with DFlash2 and FP8 MLA, ordinary generation, and five strict
-tool-call scenarios. Those exact prompts, expected behavior, actual calls, and
-raw reports are added to
-`quantization/vllm-prompt-expected-actual.md` and
-`quantization/vllm-tool-eval.json` when runtime qualification completes.
 ## Serving

 provenance and the validation report ship with the model; the full error ledger
 is retained separately as internal quantization evidence.
+## Qualified two-GPU serving result
+Release `v0.7.0` of the matching B12x/vLLM recipe qualified the public target
+revision with DFlash2 K5, FP8 MLA, TP2 + EP2 + DCP2, vision up to 16 images,
+and a 1,048,576-token request limit on 2× RTX PRO 6000 Blackwell at a 400 W
+power cap per GPU.
+- Code-agent decode: **213 tok/s at C1** and **832 tok/s at C16**.
+- 128K cold prefill: **4,572 tok/s**.
+- Available KV cache: **2,758,919 tokens**.
+- Exact 1M multi-needle test: **6/6 needles recovered**.
+- Exact 69-case thinking tool-call suite at parallelism 8: **86/100**
+  (55 pass, 8 partial, 6 fail), versus 88/100 for the matched uniform-K3
+  control.
+The full prompts, outputs, scoring, performance curves, and machine-readable
+receipts are in the recipe's
+[`v0.7.0` benchmark report](https://github.com/tpurtell/glm-5.3-flash-ext3-4-bit-2x-rtx/blob/v0.7.0/benchmarks/RESULTS.md).
 ## Serving