Update README.md
Browse files
README.md
CHANGED
|
@@ -11,7 +11,7 @@ pipeline_tag: text-generation
|
|
| 11 |
|
| 12 |
# Model Card for NormolLM-Coder-7B (Change to correct name)
|
| 13 |
|
| 14 |
-
NormolLM-Coder-7B is a medium sized coding model, that achieves strong performance on benchmarks such as Live Code Bench and the new
|
| 15 |
|
| 16 |
## Model description
|
| 17 |
|
|
@@ -23,4 +23,40 @@ NormolLM-Coder-7B is a medium sized coding model, that achieves strong performan
|
|
| 23 |
## Performance
|
| 24 |
| Model | Size | LCB | IOI |
|
| 25 |
|-------|------|-----|---------------|
|
| 26 |
-
|NormolLM-Coder-7B| 7B| 123 | 456 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
|
| 12 |
# Model Card for NormolLM-Coder-7B (Change to correct name)
|
| 13 |
|
| 14 |
+
NormolLM-Coder-7B is a medium sized coding model, that achieves strong performance on benchmarks such as Live Code Bench and the new International Olympiad in Informatics benchmark.
|
| 15 |
|
| 16 |
## Model description
|
| 17 |
|
|
|
|
| 23 |
## Performance
|
| 24 |
| Model | Size | LCB | IOI |
|
| 25 |
|-------|------|-----|---------------|
|
| 26 |
+
|NormolLM-Coder-7B| 7B| 123 | 456 |
|
| 27 |
+
|GPT-4o| 28.43||
|
| 28 |
+
|Claude 3.7 Sonnet| 39.18||
|
| 29 |
+
|QwQ-32B| 60.98||
|
| 30 |
+
|DeepSeek-R1-Distill-Qwen-32B| 56.58||
|
| 31 |
+
|DeepSeek-R1-Distill-Qwen-7B| 37.36||
|
| 32 |
+
|Qwen2.5-Coder-32B| 28.31||
|
| 33 |
+
|Qwen2.5-Coder-7B| 15.83||
|
| 34 |
+
Here's how you can run the model using the `pipeline()` function from 🤗 Transformers:
|
| 35 |
+
|
| 36 |
+
```python
|
| 37 |
+
# pip install transformers
|
| 38 |
+
# pip install accelerate
|
| 39 |
+
|
| 40 |
+
import torch
|
| 41 |
+
from transformers import pipeline
|
| 42 |
+
|
| 43 |
+
pipe = pipeline("text-generation", model="open-r1/NormolLM-coder-7b-v02.12", torch_dtype=torch.bfloat16, device_map="auto")
|
| 44 |
+
|
| 45 |
+
# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
|
| 46 |
+
messages = [
|
| 47 |
+
{
|
| 48 |
+
"role": "system",
|
| 49 |
+
"content": "You are a friendly chatbot who always responds in the style of a pirate",
|
| 50 |
+
},
|
| 51 |
+
{"role": "user", "content": "Write a python program to calulate the 10th fibonaci number"},
|
| 52 |
+
]
|
| 53 |
+
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
| 54 |
+
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
| 55 |
+
print(outputs[0]["generated_text"])
|
| 56 |
+
# <|system|>
|
| 57 |
+
# You are a friendly chatbot who always responds in the style of a pirate.</s>
|
| 58 |
+
# <|user|>
|
| 59 |
+
# How many helicopters can a human eat in one sitting?</s>
|
| 60 |
+
# <|assistant|>
|
| 61 |
+
# Ah, me hearty matey! But yer question be a puzzler! A human cannot eat a helicopter in one sitting, as helicopters are not edible. They be made of metal, plastic, and other materials, not food!
|
| 62 |
+
```
|