Text Generation
Transformers
Safetensors
qwen2
coder
code
agent
conversational
text-generation-inference
Instructions to use AdminReal/NexusCoder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AdminReal/NexusCoder with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="AdminReal/NexusCoder") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("AdminReal/NexusCoder") model = AutoModelForCausalLM.from_pretrained("AdminReal/NexusCoder", device_map="auto") 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 Settings
- vLLM
How to use AdminReal/NexusCoder with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "AdminReal/NexusCoder" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AdminReal/NexusCoder", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/AdminReal/NexusCoder
- SGLang
How to use AdminReal/NexusCoder 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 "AdminReal/NexusCoder" \ --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": "AdminReal/NexusCoder", "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 "AdminReal/NexusCoder" \ --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": "AdminReal/NexusCoder", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use AdminReal/NexusCoder with Docker Model Runner:
docker model run hf.co/AdminReal/NexusCoder
File size: 5,662 Bytes
eca5751 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 | """Code Minification Skill - Minify JS/CSS/HTML/JSON.
Strategy: remove whitespace + comments, mangle identifiers, collapse dead
code, tree-shake unused exports. Sử dụng tool phù hợp per language.
Author: Hieu Louis (2026)
"""
from __future__ import annotations
from typing import Dict, List
from .base import Skill, SkillContext, SkillCategory, SkillPriority, SkillResult
class CodeMinificationSkill(Skill):
"""Minify code: JS, CSS, HTML, JSON. Giảm size, giữ semantic."""
category = SkillCategory.CODE
priority = SkillPriority.LOW
keywords: List[str] = [
"minify", "minification", "compress code", "uglify",
"terser", "cssnano", "html minify", "nén code",
"shrink", "reduce size", "bundle size", "tree shake",
]
examples = [
"Minify this JavaScript file",
"Compress this CSS to production size",
"Uglify this JS preserving function names",
]
@property
def name(self) -> str:
return "code_minification"
@property
def description(self) -> str:
return (
"Minify JS/CSS/HTML/JSON: remove comments + whitespace, "
"mangle identifiers, collapse dead code, tree-shake unused exports."
)
def can_handle(self, prompt: str, context: SkillContext = None) -> float:
prompt_lower = prompt.lower()
score = 0.0
for kw in self.keywords:
if kw in prompt_lower:
score += 0.18
return min(1.0, score)
def execute(self, context: SkillContext) -> SkillResult:
lang = (context.language or "javascript").lower()
return SkillResult(
success=True,
output=f"[CodeMinification/{lang}] Minification strategy + toolchain ready.",
artifacts=[
{"path": "minify/strategy.md", "content": _MINIFY_STRATEGY},
{"path": "minify/example.txt", "content": _EXAMPLE_MINIFIED_JS},
],
metadata={
"skill": self.name,
"language": lang,
"toolchain": {
"javascript": "terser --compress --mangle",
"typescript": "tsc + terser (or esbuild)",
"css": "cssnano / lightningcss (Rust, fastest)",
"html": "html-minifier-terser",
"json": "jq -c (lossless)",
"python": "pyminifier (limited) — prefer zipapp + bytecode-only distribution",
},
"techniques": [
"Whitespace removal (spaces, newlines, indentation)",
"Comment stripping (// and /* */ and <!-- -->)",
"Identifier mangling (shorter names: myVar -> a)",
"Dead code elimination (unreachable statements)",
"Tree shaking (drop unused exports)",
"Constant folding (2+3 -> 5)",
"Property mangling (only when --mangle-props)",
"Hex/octal/unicode escape compression",
"Boolean shortcut (true -> !0, false -> !1)",
],
"trade_offs": {
"size_vs_debuggability": "mangled names break stack traces — ship sourcemaps to Sentry",
"size_vs_startup": "esbuild may produce slightly larger bundle but parses faster",
"compression_vs_safety": "property mangling risky with bracket access",
},
"best_practices": [
"Always emit sourcemaps (.map) and upload to error tracker",
"Measure gzipped + brotli sizes, not raw bytes",
"Use same minifier across build matrix to keep sourcemaps consistent",
"Cache bust with content-hash filenames",
],
},
suggestions=[
"Specify if source maps should be emitted",
"Indicate if identifier mangling is safe (no eval, no bracket access)",
"Check bundle budget (e.g. < 200 KB gzipped initial)",
],
)
_MINIFY_STRATEGY = """# Minification Strategy
## Per-Language Pipeline
### JavaScript / TypeScript
```bash
# terser CLI
terser input.js \\\\
--compress passes=2,drop_console=true,drop_debugger=true \\\\
--mangle toplevel \\\\
--source-map url='out.js.map' \\\\
--output out.js
```
### CSS
```bash
# lightningcss (Rust, fastest)
lightningcss --minify --bundle --targets 'defaults' input.css -o out.css
```
### HTML
```bash
html-minifier-terser \\\\
--collapse-whitespace --remove-comments \\\\
--minify-css true --minify-js true \\\\
input.html -o out.html
```
### JSON (config files)
```bash
jq -c . input.json > out.min.json
```
## Bundle Budget
- JS initial: < 200 KB gzipped
- CSS initial: < 50 KB gzipped
- Per-route lazy chunk: < 50 KB gzipped
## Pitfalls
- Mangled property names break `obj['dynamicProp']` access.
- Drop `console.log` only in production — keep in staging for tracing.
- Inline `<script>` minified by HTML minifier may conflict with CSP nonces.
"""
_EXAMPLE_MINIFIED_JS = """// ---- Before ----
function calculateTotal(items, taxRate) {
// Sum item prices
let subtotal = 0;
for (let i = 0; i < items.length; i++) {
subtotal += items[i].price * items[i].quantity;
}
const tax = subtotal * taxRate;
return subtotal + tax;
}
// ---- After (terser --compress --mangle) ----
function calculateTotal(a,b){let c=0;for(let d=0;d<a.length;d++)c+=a[d].price*a[d].quantity;return c+c*b}
// Size: 274 -> 119 bytes (-57%). Gzipped: 158 -> 96 bytes (-39%).
"""
|