Instructions to use ProCreations/grug-27b-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ProCreations/grug-27b-v2 with Transformers:

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

pipe = pipeline("text-generation", model="ProCreations/grug-27b-v2")
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)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("ProCreations/grug-27b-v2")
model = AutoModelForMultimodalLM.from_pretrained("ProCreations/grug-27b-v2", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ProCreations/grug-27b-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ProCreations/grug-27b-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ProCreations/grug-27b-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ProCreations/grug-27b-v2

SGLang

How to use ProCreations/grug-27b-v2 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 "ProCreations/grug-27b-v2" \
    --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": "ProCreations/grug-27b-v2",
		"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 "ProCreations/grug-27b-v2" \
        --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": "ProCreations/grug-27b-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ProCreations/grug-27b-v2 with Docker Model Runner:
```
docker model run hf.co/ProCreations/grug-27b-v2
```

grug-27b-v2 / release_provenance.json

ProCreations

Release validated Grug 27B v2 with integrated MTP and measured results

f572b85 verified 20 days ago

Raw History Blame Contribute Delete

6.04 kB

	{
	"release_date": "2026-09-12",
	"release_model": "ProCreations/grug-27b-v2",
	"release_gguf": "ProCreations/grug-27b-v2-gguf",
	"validated_model_revision": "79a2b2b81bf6ee0123da02467cf187db04e03708",
	"validated_gguf_revision": "30dddb92c1d580cdd966d4986398bdc69106fb08",
	"foundation_model": "Qwen/Qwen3.8-27B",
	"foundation_revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
	"v11_baseline_revision": "3ab073b4bb06dc8a83e819a33485469f32b945ba",
	"pre_defaults_update_revision": "feceb9810535346ba806a7974fa4eb71012ccded",
	"revision_equivalence": "The final defaults commit changes generation/provenance JSON only; all 13 final weight shards and the chat template match the integrated checkpoint used with explicit release settings.",
	"weight_files": {
	"model-00001-of-00012.safetensors": {
	"bytes": 2542796928,
	"sha256": "54d83c1d36631de231876217a8e0c2483eccee8746369a482b79442bdfc5d958"
	},
	"model-00002-of-00012.safetensors": {
	"bytes": 4842451920,
	"sha256": "cafff3e29f36f403fbc52bb9b50ba5d3cc8a2b40bd8c16e174a7489271d926d8"
	},
	"model-00003-of-00012.safetensors": {
	"bytes": 4965227944,
	"sha256": "f1ce55f3951686f49b281729ec4c9737035e66cc046421029e9127d9c234cf4a"
	},
	"model-00004-of-00012.safetensors": {
	"bytes": 4912819264,
	"sha256": "2a47a21efc00963f6abba1d6b450b5f5817264c83d3bdbf99f0c9e0fde2950c6"
	},
	"model-00005-of-00012.safetensors": {
	"bytes": 4986198544,
	"sha256": "c617387ac96b66a64cf1f4980ec56666dd8be7b82fc4e5c367c928e604af2527"
	},
	"model-00006-of-00012.safetensors": {
	"bytes": 4912819320,
	"sha256": "96101ea54654358d43fe5be85c625b5220b07b552c59bb165851f28b705265a5"
	},
	"model-00007-of-00012.safetensors": {
	"bytes": 4932703272,
	"sha256": "f034b71f804e4b1c868e50834dbb4921ccc4b6ca3284fbe02b6ada0c14699ec2"
	},
	"model-00008-of-00012.safetensors": {
	"bytes": 4966314576,
	"sha256": "ab6a0963f547c273ea6c920217097e31b2e4748bab2fd1b8dc2b5c697a49def8"
	},
	"model-00009-of-00012.safetensors": {
	"bytes": 4964162248,
	"sha256": "4ac90f3d606b3d1a12b9be373fff9030517cb7bb0833b7f3fbe73a4bca80d93e"
	},
	"model-00010-of-00012.safetensors": {
	"bytes": 4933789824,
	"sha256": "7102880909943841f05395c6a2dcc8e64360ca073919791c538937e4d3de5204"
	},
	"model-00011-of-00012.safetensors": {
	"bytes": 4965228032,
	"sha256": "d7a14e139b775f9d1f7742c06b316339db09a6c01cba4399c65fe95dcebaf6ff"
	},
	"model-00012-of-00012.safetensors": {
	"bytes": 2789094896,
	"sha256": "f9a1461bf3e4e5fcf7dc428dc2cae7b6a2f7d905ed83c840a8e1efd18644eaf0"
	},
	"model-mtp.safetensors": {
	"bytes": 849400424,
	"sha256": "9c59de33e8910ecd52bfd6e4e838e7062579b8e6e8b5af7c07181de2aa241c61"
	}
	},
	"final_coding_rescore_code_revision": "9ab2044b4cc3b5291a8c3da9854be3d8f71ad467",
	"evaluation_source_revisions": "See the pinned SOURCE_CODE_REVISION for each phase in results/hf_job_provenance.json; source revisions differ across phases.",
	"public_source_sha256": {
	"evaluation/agentic_eval.py": "168085e99f064ca0d2ea0427bcd180602c79eb7ac2ca6b1f0b65684691f564bd",
	"evaluation/audit_bfcl_completion.py": "bc37b5f72bc0e8e8bb01224d72652d4813d02a1d7bf7be18d442cd78fa021efc",
	"evaluation/audit_swe_commands.py": "46bb6cd75a0764a538c9b766b739d4f0cbdf1ca0846a045ce94da2ccc73ff759",
	"evaluation/bfcl_eval_subset.py": "a21f58da946bba4b38f711d84efaf65f4ccc4557d0c27e9c7f10d9e84af80121",
	"evaluation/build_gguf.py": "9c149a053e2dea261c398cc5a61a5325917dcff3699f1452544b30de85774937",
	"evaluation/coding_completion.py": "48b3203a47ecb12a4d41d67d28134d09199719dee88d7f5e968a04923c9d831e",
	"evaluation/gguf_smoke.py": "e276bf34a424b9c9ad09b77d6eeaba53955a6b41c2e94c441784a8d39c48e999",
	"evaluation/hub_io.py": "4bd7818d16356bc9830e52e154a781089113d1652b661a1d2b01ff8f26f8f952",
	"evaluation/humaneval_support.py": "aadea358e228ce4676a6bc2eba056299613f250846551c15692d6345499006aa",
	"evaluation/legacy_eval.py": "f8989e48a16dec00d931725c69fbf325199c8d75b3a9808c05c6bd30da4a631b",
	"evaluation/mtp_smoke.py": "69d2dd8483a79ba20770fc4764eb7b1d06ef818a493d25823eb10ea307ad3ead",
	"evaluation/prepare_swe.py": "edbf2c71cf83fd9f6ba4ca96ab078c8bf6d35168f864903aeb6adb2ef04aac37",
	"evaluation/repeat_audit.py": "9690dcb1dece66d93d3cc26ad48bf0bb5e31ed5031be6d1cac117dbc8887987a",
	"evaluation/repo_tasks.py": "eaf95b8a561d6250406f656d119db4e8446b576f371763786d34cd84c6debd8c",
	"evaluation/rescore_coding.py": "98bd44aee2a36bac2754a528f09a0225cefadee06347cde48c47805253d9cbec",
	"evaluation/rescore_math.py": "695e79e8d351588614d23efab8b7232029664677f642e16d7c765c40bb4892ad",
	"evaluation/run_eval.py": "4a4ac4c0a8459570257f84fe4c086818085737ffef24aebdbd635951af64c1ae",
	"evaluation/run_swe.py": "97ff00c4b10515c8733149a5c4a1e5c72576ab0900237df0788c6bc559b599f3",
	"evaluation/serving_smoke.py": "1fb7a8d1c311582c9cb52a7e60bb1e974a793e12e6cc7e4f203ef7b41198f41c",
	"evaluation/swe_common.py": "dd471175a0d9aaedec5a9c1171179a6e427689b015fcb9c25622ba2fc85ce2f7",
	"evaluation/template_v2.py": "c31aa11de5097de6310bb3b04ab88582c7e3e1653449f71d14c5cfef284f144b",
	"evaluation/verify_code.py": "f07f354fd891e9dedffdc25e16c06b46644dfc39e383af27d3be2da773eda9ff",
	"evaluation/wait_rescore_math.py": "c32748c08e7a551721268e99e5a48a2da1fbb12f38faada0bc98aeaf6acb3975",
	"normalize_messages.py": "5597ebe63e0e5911743ad723e648dacecf371651b8f28f5efc83cd2602322432",
	"training/merge_model.py": "b22f32e316b9e7218c802332ef4f973e83b82cff17467322f751340f4c435207",
	"training/tokenize_training.py": "de6d3f8fc49c310095f0781ca3c0e79325e69d00fbfae5c35d1344b4315f305f",
	"training/train_lora.py": "bdf2d1f6a5fb43fbb29767e3fcd48f257cc0acccf3752e4d6c1683b120028a38",
	"training/train_mtp.py": "075cdc9bf6173f35165351eef63e2056b430431f611a8570eaf8a1cdc4099ad7"
	},
	"local_model_compute_used": false,
	"compute_provider": "Hugging Face Jobs"
	}