Instructions to use a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed with Transformers:

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

pipe = pipeline("text-generation", model="a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed")
model = AutoModelForCausalLM.from_pretrained("a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed")
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

vLLM

How to use a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed

SGLang

How to use a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed 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 "a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed" \
    --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": "a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed",
		"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 "a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed" \
        --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": "a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed with Docker Model Runner:
```
docker model run hf.co/a-F1/Qwen2.5-Math-1.5B-Open-R1-Distill-mixed
```

a-F1 commited on Feb 24, 2025

Commit

7ba2018

verified ·

1 Parent(s): 6497b5b

End of training

Browse files

Files changed (4) hide show

README.md +3 -1
all_results.json +8 -8
config.json +1 -1
eval_results.json +8 -8

README.md CHANGED Viewed

@@ -1,9 +1,11 @@
 ---
 base_model: Qwen/Qwen2.5-Math-1.5B
 library_name: transformers
 model_name: Qwen2.5-Math-1.5B-Open-R1-Distill-mixed
 tags:
 - generated_from_trainer
 - trl
 - sft
 licence: license
@@ -11,7 +13,7 @@ licence: license
 # Model Card for Qwen2.5-Math-1.5B-Open-R1-Distill-mixed
-This model is a fine-tuned version of [Qwen/Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Quick start

 ---
 base_model: Qwen/Qwen2.5-Math-1.5B
+datasets: HuggingFaceH4/Bespoke-Stratos-17k
 library_name: transformers
 model_name: Qwen2.5-Math-1.5B-Open-R1-Distill-mixed
 tags:
 - generated_from_trainer
+- open-r1
 - trl
 - sft
 licence: license
 # Model Card for Qwen2.5-Math-1.5B-Open-R1-Distill-mixed
+This model is a fine-tuned version of [Qwen/Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) on the [HuggingFaceH4/Bespoke-Stratos-17k](https://huggingface.co/datasets/HuggingFaceH4/Bespoke-Stratos-17k) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Quick start

all_results.json CHANGED Viewed

@@ -1,13 +1,13 @@
 {
-    "eval_reasoning_loss": 0.829900324344635,
-    "eval_reasoning_runtime": 29.3011,
-    "eval_reasoning_samples_per_second": 4.403,
-    "eval_reasoning_steps_per_second": 1.126,
     "eval_samples": 100,
-    "eval_utility_loss": 1.154617190361023,
-    "eval_utility_runtime": 11.0259,
-    "eval_utility_samples_per_second": 4.444,
-    "eval_utility_steps_per_second": 1.179,
     "total_flos": 0.0,
     "train_loss": 2.266656749783226,
     "train_runtime": 29910.3238,

 {
+    "eval_reasoning_loss": 0.8222787380218506,
+    "eval_reasoning_runtime": 27.3158,
+    "eval_reasoning_samples_per_second": 4.723,
+    "eval_reasoning_steps_per_second": 1.208,
     "eval_samples": 100,
+    "eval_utility_loss": 1.4264225959777832,
+    "eval_utility_runtime": 10.2998,
+    "eval_utility_samples_per_second": 4.757,
+    "eval_utility_steps_per_second": 1.262,
     "total_flos": 0.0,
     "train_loss": 2.266656749783226,
     "train_runtime": 29910.3238,

config.json CHANGED Viewed

@@ -23,7 +23,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.50.0.dev0",
-  "use_cache": false,
   "use_mrope": false,
   "use_sliding_window": false,
   "vocab_size": 151936

   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.50.0.dev0",
+  "use_cache": true,
   "use_mrope": false,
   "use_sliding_window": false,
   "vocab_size": 151936

eval_results.json CHANGED Viewed

@@ -1,11 +1,11 @@
 {
-    "eval_reasoning_loss": 0.829900324344635,
-    "eval_reasoning_runtime": 29.3011,
-    "eval_reasoning_samples_per_second": 4.403,
-    "eval_reasoning_steps_per_second": 1.126,
     "eval_samples": 100,
-    "eval_utility_loss": 1.154617190361023,
-    "eval_utility_runtime": 11.0259,
-    "eval_utility_samples_per_second": 4.444,
-    "eval_utility_steps_per_second": 1.179
 }

 {
+    "eval_reasoning_loss": 0.8222787380218506,
+    "eval_reasoning_runtime": 27.3158,
+    "eval_reasoning_samples_per_second": 4.723,
+    "eval_reasoning_steps_per_second": 1.208,
     "eval_samples": 100,
+    "eval_utility_loss": 1.4264225959777832,
+    "eval_utility_runtime": 10.2998,
+    "eval_utility_samples_per_second": 4.757,
+    "eval_utility_steps_per_second": 1.262
 }