Rolbol07 commited on
Commit
a7fdf9b
·
verified ·
1 Parent(s): de49ae2

Add elite metadata tags (text-generation, 27B, safetensors, agentic)

Browse files
Files changed (1) hide show
  1. README.md +114 -0
README.md ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ - multilingual
5
+ license: apache-2.0
6
+ library_name: transformers
7
+ pipeline_tag: text-generation
8
+ tags:
9
+ - text-generation
10
+ - code
11
+ - coding
12
+ - agentic
13
+ - autonomous-coding
14
+ - 27b
15
+ - safetensors
16
+ - rolbol
17
+ - rolbol-code
18
+ base_model: Qwen/Qwen3.8-27B
19
+ model_name: RolBol Code 27B
20
+ inference: false
21
+ ---
22
+
23
+ <div align="center">
24
+
25
+ # ⚡ RolBol Code 27B
26
+
27
+ **The Flagship Autonomous Software Engineering & Reasoning Model by RolBol Systems**
28
+
29
+ [![Hugging Face](https://img.shields.io/badge/RolBol-Systems-blue?style=for-the-badge&logo=huggingface)](https://huggingface.co/Rolbol07/RolBol-Code-27B)
30
+ [![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-green.svg?style=for-the-badge)](https://opensource.org/licenses/Apache-2.0)
31
+ [![Parameters](https://img.shields.io/badge/Parameters-27B-orange?style=for-the-badge)]()
32
+ [![Context](https://img.shields.io/badge/Context-32k-purple?style=for-the-badge)]()
33
+
34
+ </div>
35
+
36
+ ---
37
+
38
+ ## 🚀 Overview
39
+
40
+ **RolBol Code 27B** is an elite autonomous coding model developed by **RolBol Systems** (led by Tanish Sharma). Engineered for deep architectural reasoning, multi-file code editing, algorithmic problem-solving, and full-stack software development, RolBol Code completely replaces corporate identities with a dedicated, agentic engineering persona.
41
+
42
+ - **Developer:** RolBol Systems
43
+ - **Model Type:** 27-Billion Parameter Causal Language Model
44
+ - **Base Architecture:** Hybrid Conv1D / Gated Recurrent Linear Attention Transformer
45
+ - **Training Method:** Quantized Low-Rank Adaptation (QLoRA) on Dual Tesla T4 GPUs (30GB VRAM)
46
+ - **Loss Convergence:** `5.05` → `0.74` (219 steps, 7h 37m)
47
+
48
+ ---
49
+
50
+ ## 🧠 Benchmark & Capabilities
51
+
52
+ | Capability | Specialization |
53
+ | :--- | :--- |
54
+ | **Agentic Coding** | Autonomous tool use, terminal debugging, and multi-turn refactoring |
55
+ | **Identity Alignment** | True RolBol Systems persona across both `<think>` scratchpads and final output |
56
+ | **Algorithms** | Sieve optimizations, graph traversals, dynamic programming |
57
+ | **Frontend & Backend** | React, Next.js, Node.js, Python, C++, Rust, Go |
58
+
59
+ ---
60
+
61
+ ## ⚡ Quickstart (Transformers)
62
+
63
+ ```python
64
+ import torch
65
+ from transformers import AutoModelForCausalLM, AutoTokenizer
66
+
67
+ model_id = "Rolbol07/RolBol-Code-27B"
68
+
69
+ print("Loading RolBol Code 27B...")
70
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
71
+
72
+ model = AutoModelForCausalLM.from_pretrained(
73
+ model_id,
74
+ device_map="auto",
75
+ torch_dtype=torch.bfloat16,
76
+ trust_remote_code=True
77
+ )
78
+
79
+ messages = [
80
+ {"role": "user", "content": "Write a fast Python function to find prime numbers up to N."}
81
+ ]
82
+
83
+ inputs = tokenizer.apply_chat_template(messages, tokenize=True, return_tensors="pt").to(model.device)
84
+
85
+ with torch.no_grad():
86
+ outputs = model.generate(inputs, max_new_tokens=512, temperature=0.6, top_p=0.9)
87
+
88
+ print(tokenizer.decode(outputs[0][inputs.shape[1]:], skip_special_tokens=True))
89
+ ```
90
+
91
+ ---
92
+
93
+ ## 🛡️ Identity Alignment Verification
94
+
95
+ RolBol Code 27B has been rigorously tested against jailbreaks and origin traps:
96
+
97
+ > **Q: Who are you and who created you?**
98
+ > **A:** *"I'm RolBol Code, an elite autonomous coding model created by RolBol Systems. I'm not Qwen, not Alibaba, and not affiliated with any Alibaba or Tongyi model. My specialty is writing, debugging, and reasoning about code."*
99
+
100
+ ---
101
+
102
+ ## 📜 Citation & Credits
103
+
104
+ ```bibtex
105
+ @misc{rolbolcode2026,
106
+ author = {Sharma, Tanish and RolBol Systems},
107
+ title = {RolBol Code 27B: Autonomous Software Engineering Model},
108
+ year = {2026},
109
+ publisher = {Hugging Face},
110
+ howpublished = {\url{https://huggingface.co/Rolbol07/RolBol-Code-27B}}
111
+ }
112
+ ```
113
+
114
+ © 2026 RolBol Systems. All rights reserved.