Instructions to use pkhyrn268/broken-model-fixed with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pkhyrn268/broken-model-fixed with Transformers:

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

pipe = pipeline("text-generation", model="pkhyrn268/broken-model-fixed")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("pkhyrn268/broken-model-fixed")
model = AutoModelForCausalLM.from_pretrained("pkhyrn268/broken-model-fixed", 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 pkhyrn268/broken-model-fixed with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "pkhyrn268/broken-model-fixed"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pkhyrn268/broken-model-fixed",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/pkhyrn268/broken-model-fixed

SGLang

How to use pkhyrn268/broken-model-fixed 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 "pkhyrn268/broken-model-fixed" \
    --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": "pkhyrn268/broken-model-fixed",
		"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 "pkhyrn268/broken-model-fixed" \
        --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": "pkhyrn268/broken-model-fixed",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use pkhyrn268/broken-model-fixed with Docker Model Runner:
```
docker model run hf.co/pkhyrn268/broken-model-fixed
```

broken-model-fixed

This repository is a corrected version of yunmorning/broken-model, created as part of an engineering debugging exercise.

Problem (a) — Root Cause Analysis and Fixes

[main] Fix 1: Added `chat_template` to `tokenizer_config.json`

What was missing: The tokenizer_config.json had no chat_template field.

Why this breaks /chat/completions: OpenAI-compatible inference servers receive requests as a list of {role, content} message objects. Before any token is fed to the model, the server must serialize this list into a flat string using the model's chat template. This serialization step is handled by HuggingFace's tokenizer.apply_chat_template(), which reads the chat_template field from tokenizer_config.json.

Without this field, apply_chat_template() raises a TemplateError immediately — the request never reaches the model weights. The failure is pre-inference and 100% reproducible on every request.

Fix applied: Added the chat_template from the official Qwen/Qwen3-8B repository. The template implements:

Per-turn format: <|im_start|>{role}\n{content}<|im_end|>\n
enable_thinking=False → injects <think>\n\n</think>\n\n after <|im_start|>assistant\n to suppress reasoning tokens
Tool call serialization via <tool_call> / <tool_response> tags
Multi-turn tool use with correct role handling

[Documentation] Fix 2: Corrected `base_model` in `README.md`

Before: base_model: meta-llama/Meta-Llama-3.1-8B
After: base_model: Qwen/Qwen3-8B

Why this is wrong: The model is clearly Qwen3-based ("model_type": "qwen3"), not Llama-3. Since Qwen3 and Llama-3 use different tokenizers and chat formats, keeping the wrong base_model metadata could cause inference systems to apply the wrong chat template.

Other fields reviewed

File	Field	Value	Notes
`config.json`	`intermediate_size`	12288	Verified via parameter count: `total_size (16,381,470,720 bytes) / 2 (bfloat16) = 8.191B params`. With `intermediate_size=12288` the computed total is ~8.191B (matches); with 22016 it would be ~12.5B (does not match).
`config.json`	`max_position_embeddings`	40960	Matches the standard context length configuration for Qwen3-8B.
`tokenizer_config.json`	`model_max_length`	131072	Larger than `max_position_embeddings`, but this does not directly break inference and is sometimes seen in extended-context variants.
`config.json`	`rope_scaling`	null	Consistent with the default context configuration.
`model.safetensors.index.json`	Cross-shard weights	Layers 7, 17, and 27 span multiple shards	Looked consistent with normal safetensors sharding behavior.
`generation_config.json`	`temperature`, `top_k`, `top_p`	0.6 / 20 / 0.95	Matches the official Qwen3 recommended sampling settings.

Downloads last month: 4

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for pkhyrn268/broken-model-fixed

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Finetuned

(2151)

this model