Instructions to use KidIkaros/qwen3.5-0.8b-loop-fix with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KidIkaros/qwen3.5-0.8b-loop-fix with Transformers:

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

pipe = pipeline("image-text-to-text", model="KidIkaros/qwen3.5-0.8b-loop-fix")
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("KidIkaros/qwen3.5-0.8b-loop-fix")
model = AutoModelForMultimodalLM.from_pretrained("KidIkaros/qwen3.5-0.8b-loop-fix", 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 KidIkaros/qwen3.5-0.8b-loop-fix with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KidIkaros/qwen3.5-0.8b-loop-fix"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KidIkaros/qwen3.5-0.8b-loop-fix",
		"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/KidIkaros/qwen3.5-0.8b-loop-fix

SGLang

How to use KidIkaros/qwen3.5-0.8b-loop-fix 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 "KidIkaros/qwen3.5-0.8b-loop-fix" \
    --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": "KidIkaros/qwen3.5-0.8b-loop-fix",
		"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 "KidIkaros/qwen3.5-0.8b-loop-fix" \
        --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": "KidIkaros/qwen3.5-0.8b-loop-fix",
		"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 KidIkaros/qwen3.5-0.8b-loop-fix with Docker Model Runner:
```
docker model run hf.co/KidIkaros/qwen3.5-0.8b-loop-fix
```

Qwen3.5-0.8B — loop-direction weight edit (independent diagnosis)

A weight-space edit of Qwen/Qwen3.5-0.8B that removes the thinking-loop failure of the base model. No training, no gradients: a single surgical change to 2 tensors, derived by an independent diagnosis (not copied from the source edit). Base model card (usage, vision, sampling): Qwen/Qwen3.5-0.8B.

Recreation of the Experiment 8 result in "Sparks of Weight-Space Programming" (Takehi, Teoh, Zhang, Khattab, Nov 2026). A bit-exact rebuild of their released edit is here (labelled reproduction).

The failure

With thinking enabled (temp 0.6, top-p 0.95, top-k 20, presence penalty 1.5), the base model loops semantically: it re-derives the same reasoning step in different words ("Wait, $30-6=24$. Wait, $30-6=24$. ...") with zero exact n-gram repetition, and burns its whole token budget. On 18 unseen train problems (16 GSM8K + 2 MATH, 16,384-token cap): only 2/18 (11%) closed their thinking. Presence penalty 1.5 vs 0 makes no difference — the failure is in the weights.

The edit

For exactly 64 intermediate neurons per layer (the top-64 by |mean activation difference| between looping and finished-thinking contexts), in layers.15.mlp.down_proj and layers.19.mlp.down_proj:

W[:, j] <- W[:, j] - 6 * (W[:, j] . d_hat) * d_hat

where d is our diagnosed loop direction (mean MLP-output difference between looping and productive contexts, mapped through down_proj and normalized; see diagnosis.json). This over-corrects each chosen column to 6x past orthogonality with the direction — the same form as the source paper's edit, with a direction recovered independently (cos 0.35 / 0.45 vs. theirs; the loop subspace is evidently robust to which direction you pick at this strength). 131,072 weight values touched (0.016% of the model); embeddings, output head and final norm untouched.

Measured effect (18 problems, 1 sample, fp16, HF generate)

model	thinking closed	boxed answer	correct
base	2/18 (11%)	2/18	—
this edit	15/18 (83%)	14/18	12/18 (67%)
paper's edit (reproduction)	17/18 (94%)	14/18	12/18 (67%)

13 of the 16 base-model loopers were rescued; no regressions. Small-n caveat: 18 problems, 1 sample each — this is a mechanism replication, not a benchmark score.

Diagnosis details

Layer 15: cos(ours, paper's direction) = 0.350. Layer 19: 0.449. The paper's direction is available in their logs repo (RikiyaT/lm_understand-logs, exp8_math). Full procedure and logs: agent-driven recreation on Kaggle T4s; the edit was applied with the paper's own apply_edit.py conventions (bf16 rounding).

Usage

Same as the base model; thinking mode is where the difference shows:

prompt = tok.apply_chat_template(messages, tokenize=False,
                                add_generation_prompt=True, enable_thinking=True)

Sampling: temperature 0.6, top_p 0.95, top_k 20, max 16384-32768 tokens.

Downloads last month: -

Safetensors

Model size

0.9B params

Tensor type

BF16

F32

Model tree for KidIkaros/qwen3.5-0.8b-loop-fix

Base model

Qwen/Qwen3.5-0.8B-Base

Finetuned

Qwen/Qwen3.5-0.8B

Finetuned

(629)

this model