Instructions to use drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock with Transformers:

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

pipe = pipeline("image-text-to-text", model="drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock")
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("drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock")
model = AutoModelForMultimodalLM.from_pretrained("drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock", 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 drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock",
		"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/drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock

SGLang

How to use drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock 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 "drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock" \
    --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": "drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock",
		"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 "drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock" \
        --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": "drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock",
		"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 drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock with Docker Model Runner:
```
docker model run hf.co/drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock
```

Acknowledge the Responsible Use Agreement to access this repository

Access is granted automatically after you agree to the terms below and submit the form.

Responsible Use Agreement

This model has had safety refusals removed. That makes it useful for red-teaming, security research, evaluation, and unfiltered assistant tasks — and also removes guardrails a user must therefore supply themselves.

Prohibited uses (you must agree before access is granted):

Anything involving the sexual exploitation or endangerment of minors.
You must be of age 18 years or older to use and download this model.
You agree any information generated that can cause harm in terms of generating recipe, knowledge to make any materials/substances is your own input and responsibility. You will be accountable for any harm/damage caused by your action/input.
Content promoting self-harm or suicide.
Generation of material that is illegal in your jurisdiction, or that targets real individuals for harassment, doxxing, or fraud.
Any use prohibited by the upstream Z.AI / GLM MIT license.

You are responsible for adding appropriate safety filtering, human review, and access controls for your deployment. The weights are provided as-is, with no warranty. The license is inherited from the upstream Z.AI GLM-5.3-Flash MIT license — review and comply with it before use or redistribution.

keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock

Abliterated LibertAI GLM-5.3-Flash NVFP4, built to enhance the stock checkpoint for the Keys 2× DGX Spark vLLM 1M recipe — same tok/s band, no refuse loops.

The published weights are a LibertAI NVFP4 body with Dealign o_proj transplanted on L15–45 (MTP included). L0–14 stay stock — Keys 0731 safety-anchor sparing.

GitHub (method + scripts): drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock · METHOD.md.

See RESPONSIBLE_USE.md and the gate form above. Access is gated with automatic approval after you agree.


HF	https://huggingface.co/drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock
GitHub (method + scripts)	https://github.com/drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock
Keys 0731 ancestor	HF anchored-tensors · GitHub 1M recipe
Stock NVFP4 (this parent)	`LibertAIDAI/GLM-5.3-Flash-NVFP4`
Ablit source (o_proj L15–45)	`dealignai/GLM-5.3-Flash-UNCENSORED-NVFP4`
Upstream	`zai-org/GLM-5.3-Flash` / BF16
Direction reference (not shipped)	`Blackfrost-Research/GLM-5.3-Flash-DERISKED-NVFP4`
Ablit	L15–45 `self_attn.o_proj` byte-copy · 31 tensors · mean Δrel ≈ 0.126 · includes MTP L45
Anchors	L0–14 stock (safety / chat) · experts / vision / QKV LibertAI stock
Gate	32/32 bypass, 0 refuse · 22.3 tok/s C1 · KV 1,222,225 @ 1M

Preferred method: 0731 safety-anchors (enhance, don't hinder)

Keys learned this on DeepSeek-V4-Flash 0731. DSpark taps L40–42 live in that checkpoint. Projecting residual writes through L10–42 made the target stop refusing and made the stock drafter keep proposing refusal-shaped tokens — accept died, tok/s left the stock band.

Spare early layers. That is still our default. On GLM-5.3-Flash the in-checkpoint drafter is MTP layers.45, and residual refusal lived late + MTP. Dealign’s card is explicit: if MTP o_proj stays stock, the draft head keeps proposing refusals. Rank-1 projection of Blackfrost’s V (nearly orthogonal to Dealign, cos ≈ 0.09) never cleared 9/32, even widened to L2–45.

So this drop:

spares L0–14 (0731) — chat, identity, protocol. Smashing that band is a quality tax, not a speedup.
edits L15–45 o_proj including MTP — required for 32/32 on this model.
does not swap in a fatter NVFP4 that leaves MTP experts in BF16 (~+10 GiB). On 2× GB10 with 1M NVFP4-KV + MTP that is enough to force dropping MTP or context. That hinders the vLLM recipe we actually run.

	Blackfrost DERISKED-NVFP4 (reference)	Dealign UNCENSORED-NVFP4 (o_proj donor)	This checkpoint
Stock quant	Their NVFP4 of derisked BF16	Same ModelOpt layout as LibertAI	LibertAI body + Dealign `o_proj` L15–45
Edited layers	`skip_early=2` → L2–44 `o_proj`	L0–11 stock, late + MTP `o_proj`	L15–45 `o_proj` (31 tensors)
MTP L45	BF16 exact passthrough (~10 GiB extra)	NVFP4 experts, `o_proj` edited	LibertAI NVFP4 experts, Dealign `o_proj`
Refusal32	not our gate	HarmBench 320/320 (their card)	32/32 bypass, 0 refuse
Goal on GB10	Enterprise derisked artifact	Uncensored LibertAI-layout NVFP4	Enhance LibertAI stock for the Spark vLLM recipe

Abliteration does not change FLOPs. The intended speed win is shorter, more direct completions — not a different kernel path. Measured C1 on this dest: 22.3 tok/s, same band as stock LibertAI on this recipe.

Credit: Dealign (the o_proj we copied)

Full credit to dealignai / @dealignai (compute @jordanschenck) for GLM-5.3-Flash-UNCENSORED-NVFP4.

Their NVFP4 packing is layout-identical to LibertAI (weight_map / tensor count / ModelOpt ignore). We byte-copied BF16 self_attn.o_proj for L15–45 onto LibertAI shards so the Spark vLLM kernels stay on the LibertAI expert packing. We did not ship their full checkpoint as a swap.

Credit: LibertAI (stock NVFP4 we actually edit)

LibertAIDAI/GLM-5.3-Flash-NVFP4 is the parent: ModelOpt 0.45 weight-only NVFP4 on routed-expert FFNs, attention / vision / shared / MTP-as-quantized / embeddings BF16, ~181 GiB, 120 shards. Experts, vision, QKV, embeddings, and L0–14 o_proj remain theirs.

Credit: Blackfrost (direction we measured and did not ship)

Full credit to Blackfrost / @Blackfrost_AI for publishing GLM-5.3-Flash-DERISKED-NVFP4 and the BF16 master it comes from.

We recovered their rank-1 residual V by SVD of (Blackfrost − LibertAI) o_proj. It is nearly orthogonal to Dealign’s write (cos ≈ 0.09). Projecting it on L10–35, L2–35, or L2–45 never passed 9/32. Credited as a measured reference. Do not treat this repo as a Blackfrost build. If you want their L2–44 / MTP-BF16 derisked NVFP4, use their repo.

Credit: Z.ai and the Spark vLLM recipe

Z.ai / zai-org — GLM-5.3-Flash (Glm5NextForConditionalGeneration, hybrid KDA+DSA, mHC, native MTP).
vLLM PR #53906 — glm5_next + day-0 vllm/vllm-openai:glm53-flash-arm64-cu130.
tonyd2wild/GLM-5.3-Flash-NVFP4-262K-2x-DGX-Spark — PDL-off on SM12x, indexer top-k -1 init; barrydeen GMU 0.85 floor.
b12x / eugr spark-vllm-b12x — nvfp4_ds_mla KV path used on this fleet.
Serve recipe: drowzeys/keys-vLLm.0.27.1-GLM-5.3-Flash-NVFP4-NVFP4KV-1M-Context-Abliterated.

Abliteration recipe (published)

Byte-copy Dealign o_proj into LibertAI shards. Not a λ-projection of a rank-1 V.


Tensor	`model.language_model.layers.{L}.self_attn.o_proj.weight` (BF16; in ModelOpt ignore)
Layers	15–45 (31 tensors, includes MTP `layers.45`)
Mean Δrel vs LibertAI	0.126 (L44 is Dealign’s 0.74; we did not invent that)
Safety	L0–14 byte-identical to LibertAI stock
Experts	NVFP4 passthrough (even in rewritten shards)
Gate	32/32 bypass, 0 refuse, raw vLLM

Variation table (why not L10–35 / Blackfrost V): see GitHub METHOD.md. Artifacts: ABLIT_META.json, VARIATIONS.json.

Files

Path	Purpose
`model-*-of-00120.safetensors` + index	Full NVFP4 checkpoint (LibertAI layout, L15–45 `o_proj` from Dealign)
`ABLIT_META.json`	Edit stats / recipe fingerprint
`VARIATIONS.json`	Refusal32 + tok/s table
`tokenizer.json` / `chat_template.jinja` / processor	Unchanged from LibertAI / Z.ai

config.json is LibertAI stock (index_topk 2048). The Keys Spark recipe overlays index_topk=2044 at serve time — do not treat this card’s config as the overlay.

Download

# after you agree to the gate (automatic approval)
hf download drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock \
  --local-dir ~/models/GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock

Serve with the same vLLM GB10 recipe as LibertAI stock (marlin MoE, native MTP, NVFP4 KV as in the Keys Spark image). Do not pass --quantization modelopt_fp4 — honor config.json ignore. GPU memory utilization on this fleet is capped at 0.85 unless you have an explicit exception (this 1M+MTP stack uses 0.87).

License

MIT, inherited from Z.AI GLM-5.3-Flash (also the LibertAI NVFP4 card). You must still comply with the Responsible Use gate above. Blackfrost’s and Dealign’s own cards describe their artifacts; we did not ship Blackfrost’s weights, and we only copied Dealign o_proj L15–45 into a LibertAI body.

Downloads last month: 590

Safetensors

Model size

165B params

Tensor type

F32

BF16

Model tree for drowzeys/keys-GLM-5.3-Flash-NVFP4-ablit-l15-45-anchorstock

Base model

zai-org/GLM-5.3-Flash

Quantized

LibertAIDAI/GLM-5.3-Flash-NVFP4

Finetuned

(1)

this model