Update README.md
Browse files
README.md
CHANGED
@@ -10,11 +10,11 @@ tags:
|
|
10 |
- pytorch
|
11 |
---
|
12 |
|
13 |
-
#
|
14 |
|
15 |
## Description
|
16 |
|
17 |
-
|
18 |
It is a reasoning model that is post trained for reasoning while code generation. The model supports a context length of 32K tokens.
|
19 |
|
20 |
This model is ready for commercial use.
|
@@ -29,7 +29,7 @@ Global<br>
|
|
29 |
This model is intended for developers and researchers building LLMs. <br>
|
30 |
|
31 |
### Release Date: <br>
|
32 |
-
Huggingface [04/25/2025] via https://huggingface.co/nvidia/
|
33 |
|
34 |
|
35 |
## References
|
@@ -70,13 +70,13 @@ Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated sys
|
|
70 |
|
71 |
|
72 |
## Training Dataset
|
73 |
-
The training corpus for
|
74 |
* Data Collection Method: Hybrid: Automated, Human, Synthetic <br>
|
75 |
* Data Labeling Method: Hybrid: Automated, Human, Synthetic <br>
|
76 |
|
77 |
|
78 |
## Evaluation Dataset
|
79 |
-
We used the datasets listed in the next section to evaluate OpenCodeReasoning-
|
80 |
* Data Collection Method: Hybrid: Automated, Human, Synthetic <br>
|
81 |
* Data Labeling Method: Hybrid: Automated, Human, Synthetic <br>
|
82 |
|
|
|
10 |
- pytorch
|
11 |
---
|
12 |
|
13 |
+
# OpenCodeReasoning-Nemotron-14B Overview
|
14 |
|
15 |
## Description
|
16 |
|
17 |
+
OpenCodeReasoning-Nemotron-14B is a large language model (LLM) which is a derivative of [Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct) (AKA the *reference model*).
|
18 |
It is a reasoning model that is post trained for reasoning while code generation. The model supports a context length of 32K tokens.
|
19 |
|
20 |
This model is ready for commercial use.
|
|
|
29 |
This model is intended for developers and researchers building LLMs. <br>
|
30 |
|
31 |
### Release Date: <br>
|
32 |
+
Huggingface [04/25/2025] via https://huggingface.co/nvidia/OpenCodeReasoning-Nemotron-14B/ <br>
|
33 |
|
34 |
|
35 |
## References
|
|
|
70 |
|
71 |
|
72 |
## Training Dataset
|
73 |
+
The training corpus for OpenCodeReasoning-Nemotron-14B is [OpenCodeReasoning](https://huggingface.co/datasets/nvidia/OpenCodeReasoning) dataset, which is composed of competitive programming questions and DeepSeek-R1 generated responses.
|
74 |
* Data Collection Method: Hybrid: Automated, Human, Synthetic <br>
|
75 |
* Data Labeling Method: Hybrid: Automated, Human, Synthetic <br>
|
76 |
|
77 |
|
78 |
## Evaluation Dataset
|
79 |
+
We used the datasets listed in the next section to evaluate OpenCodeReasoning-Nemotron-14B. <br>
|
80 |
* Data Collection Method: Hybrid: Automated, Human, Synthetic <br>
|
81 |
* Data Labeling Method: Hybrid: Automated, Human, Synthetic <br>
|
82 |
|