Instructions to use kd13/Coder-o1-mini-reasoning-AWQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kd13/Coder-o1-mini-reasoning-AWQ with Transformers:

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

pipe = pipeline("text-generation", model="kd13/Coder-o1-mini-reasoning-AWQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kd13/Coder-o1-mini-reasoning-AWQ")
model = AutoModelForCausalLM.from_pretrained("kd13/Coder-o1-mini-reasoning-AWQ", 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 kd13/Coder-o1-mini-reasoning-AWQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kd13/Coder-o1-mini-reasoning-AWQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kd13/Coder-o1-mini-reasoning-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kd13/Coder-o1-mini-reasoning-AWQ

SGLang

How to use kd13/Coder-o1-mini-reasoning-AWQ 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 "kd13/Coder-o1-mini-reasoning-AWQ" \
    --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": "kd13/Coder-o1-mini-reasoning-AWQ",
		"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 "kd13/Coder-o1-mini-reasoning-AWQ" \
        --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": "kd13/Coder-o1-mini-reasoning-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kd13/Coder-o1-mini-reasoning-AWQ with Docker Model Runner:
```
docker model run hf.co/kd13/Coder-o1-mini-reasoning-AWQ
```

kd13 commited on Jul 21

Commit

92becf4

verified ·

1 Parent(s): 30b238c

Update README.md

Browse files

Files changed (1) hide show

README.md +7 -36

README.md CHANGED Viewed

@@ -12,28 +12,19 @@ tags:
 - quantized
 - vllm
 - code
-- text-generation
 ---
 # Coder-o1-mini-reasoning - AWQ
-4-bit AWQ quantization of [kd13/Coder-o1-mini-reasoning](https://huggingface.co/kd13/Coder-o1-mini-reasoning), a compact
-Python-focused reasoning model for coding assistance, debugging, code explanation, and
-math/logic reasoning.
-Quantized with [llm-compressor](https://github.com/vllm-project/llm-compressor) using
-`AWQModifier` + `W4A16_ASYM`. Calibrated on 256 code-instruction samples at
-2048 tokens, with the model's own chat template applied.
 `lm_head` is left at full precision. Weights are 4-bit; activations stay 16-bit.
 ## Format
-This is **compressed-tensors** format, which is what current AWQ tooling produces.
-vLLM and transformers both detect it automatically from `config.json` — you do not need
-to pass `--quantization awq`. The older AutoAWQ format is not interchangeable with this
-one; if a loader expects `quant_config.json`, it wants the legacy format and will not
-read this repo.
 ## Usage
@@ -64,12 +55,9 @@ Requires `pip install compressed-tensors`.
 ## Hardware
-A CUDA GPU is required — AWQ has no CPU path. For local or CPU inference use the GGUF
-build instead.
-On Ampere or newer (compute capability 8.0+) vLLM uses the Marlin kernel, which is where
-the throughput gains come from. Turing cards such as the T4 fall back to a slower kernel
-and see much less benefit.
 ## Chat template
@@ -83,23 +71,6 @@ ChatML, with Qwen-style tool calling:
 <|im_start|>assistant
 ```
-Tool definitions are injected into the system message inside `<tools>` tags, and the
-model replies with a JSON object inside `<tool_call>` tags. Tool results are returned
-wrapped in `<tool_response>`. vLLM exposes this through its OpenAI-compatible `tools`
-parameter.
-A default system prompt is applied when you do not supply one. Pass an explicit system
-prompt to control the assistant's stated identity.
-## Limitations
-Everything in the [base model card](https://huggingface.co/kd13/Coder-o1-mini-reasoning) applies. This is
-a small experimental reasoning model: good for Python learning, debugging help, code
-explanation, and basic-to-intermediate problems. Not suited to hard competitive
-programming, complex algorithmic work, non-Python languages, security-sensitive code, or
-production use without review.
-4-bit quantization does not improve any of that. Expect 1-3% degradation on most tasks,
-concentrated in exactly the long multi-step reasoning this model is already weakest at.
-If quality matters more than memory, use the unquantized model or an 8-bit build.
-Always test generated code.

 - quantized
 - vllm
 - code
 ---
 # Coder-o1-mini-reasoning - AWQ
+4-bit AWQ quantization of [kd13/Coder-o1-mini-reasoning](https://huggingface.co/kd13/Coder-o1-mini-reasoning), a compact Python-focused reasoning model for coding assistance, debugging, code explanation, and math/logic reasoning.
+Quantized with [llm-compressor](https://github.com/vllm-project/llm-compressor) using `AWQModifier` + `W4A16_ASYM`. Calibrated on 256 code-instruction samples at 2048 tokens, with the model's own chat template applied.
 `lm_head` is left at full precision. Weights are 4-bit; activations stay 16-bit.
 ## Format
+This is **compressed-tensors** format, which is what current AWQ tooling produces. vLLM and transformers both detect it automatically from `config.json` — you do not need to pass `--quantization awq`. The older AutoAWQ format is not interchangeable with this one; if a loader expects `quant_config.json`, it wants the legacy format and will not read this repo.
 ## Usage
 ## Hardware
+A CUDA GPU is required — AWQ has no CPU path. For local or CPU inference use the GGUF build instead.
+On Ampere or newer (compute capability 8.0+) vLLM uses the Marlin kernel, which is where the throughput gains come from. Turing cards such as the T4 fall back to a slower kernel and see much less benefit.
 ## Chat template
 <|im_start|>assistant
 ```
+Tool definitions are injected into the system message inside `<tools>` tags, and the model replies with a JSON object inside `<tool_call>` tags. Tool results are returned wrapped in `<tool_response>`. vLLM exposes this through its OpenAI-compatible `tools` parameter.
+A default system prompt is applied when you do not supply one. Pass an explicit system prompt to control the assistant's stated identity.