Arxiv:2403.07691 Autotrain compatible Code Conversational Endpoints compatible Qwen2 Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/NTQAI/Nxcode-CQ-7B-orpo
Nxcode CQ 7B Orpo Benchmarks
MMLU Pro: 6.79
GPQA: 0.56
MUSR: 7.05
BBH: 17.58
IFEval: 40.07 vs 88 (so35)-54.5%
ARC: 35.49 vs 96.7 (so35)-63.3%
HellaSwag: 53.86 vs 95.3 (gpt4)-43.5%
MMLU: 39.24 vs 88.3 (so35)-55.6%
TruthfulQA: 45.01 vs 59 (gpt4)-23.7%
WinoGrande: 57.22 vs 87.5 (gpt4)-34.6%
GSM8K: 27.07 vs 96.4 (so35)-71.9%
HumanEvalA: 55.24
Python Eva: 87.23 vs 92 (so35)-5.2%
Java Eval: 60.91
Javascript: 71.69
CPP Eval: 68.04
PHP Eval: 67.98
Julia Eval: 50.54
D Eval: 23.22
Lua Eval: 37.24
R Eval: 39.73
Racket Eva: 32.26
Rust Eval: 65.53
Swift Eval: 58.51
MATH Lvl 5: 2.19
LLME Score: 0.25738
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Nxcode CQ 7B Orpo Parameters and Internals
| Model Type |
|
| LLM Name | Nxcode CQ 7B Orpo |
| Repository 🤗 | https://huggingface.co/NTQAI/Nxcode-CQ-7B-orpo |
| Model Size | 7b |
| Required VRAM | 14.5 GB |
| Updated | 2025-09-16 |
| Maintainer | NTQAI |
| Model Type | qwen2 |
| Model Files | |
| Model Architecture | Qwen2ForCausalLM |
| License | other |
| Context Length | 65536 |
| Model Max Length | 65536 |
| Transformers Version | 4.39.3 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <fim_pad> |
| Vocabulary Size | 92416 |
| Torch Data Type | bfloat16 |
Best Alternatives to Nxcode CQ 7B Orpo
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Qwen2.5 7B Instruct 1M | 986K / 15.4 GB | 46883 | 346 |
| Hush Qwen2.5 7B V1.2 | 986K / 15.2 GB | 6 | 1 |
| Hush Qwen2.5 7B V1.1 | 986K / 15.2 GB | 6 | 1 |
| Hush Qwen2.5 7B V1.4 | 986K / 15.2 GB | 6 | 1 |
| Hush Qwen2.5 7B Preview | 986K / 15.2 GB | 8 | 0 |
| Hush Qwen2.5 7B V1.3 | 986K / 15.2 GB | 6 | 2 |
| Qwen2.5 7B Preview | 986K / 15.2 GB | 5 | 0 |
| Hush Qwen2.5 7B RP V1.4 1M | 986K / 15.2 GB | 5 | 2 |
| Qwen 2.5 7B Exp Sce | 986K / 15.2 GB | 5 | 2 |
| Qwen2.5 7B MixStock V0.1 | 986K / 15.2 GB | 7 | 3 |
Note: green Score (e.g. "73.2") means that the model is better than NTQAI/Nxcode-CQ-7B-orpo .
Rank the Nxcode CQ 7B Orpo Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 51408 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124