Hugging Face's logo

Datasets:
codeShare
/
lora-training-data

Tags:
Not-For-All-Audiences
Dataset card Files
xet
 Community
1
codeShare commited on
Commit
157995e
·
verified ·
1 Parent(s): 55b3d84

Upload HauauCS_Abliterated_Qwen_3_5_4B_LLM_model.ipynb

Browse files
Files changed (1) hide show
  1. HauauCS_Abliterated_Qwen_3_5_4B_LLM_model.ipynb +1 -808
HauauCS_Abliterated_Qwen_3_5_4B_LLM_model.ipynb CHANGED
@@ -1,808 +1 @@
1
- {
2
- "nbformat": 4,
3
- "nbformat_minor": 0,
4
- "metadata": {
5
- "colab": {
6
- "provenance": [],
7
- "gpuType": "T4"
8
- },
9
- "kernelspec": {
10
- "name": "python3",
11
- "display_name": "Python 3"
12
- },
13
- "language_info": {
14
- "name": "python"
15
- },
16
- "accelerator": "GPU"
17
- },
18
- "cells": [
19
- {
20
- "cell_type": "markdown",
21
- "source": [
22
- "# HauauCS Abliterated Qwen 3.5 -4B LLM model (image and/or text input)"
23
- ],
24
- "metadata": {
25
- "id": "ewVsp7wRAH2j"
26
- }
27
- },
28
- {
29
- "cell_type": "code",
30
- "execution_count": null,
31
- "metadata": {
32
- "id": "lC55DC9m_4Zz"
33
- },
34
- "outputs": [],
35
- "source": [
36
- "\n",
37
- "# ============================================================\n",
38
- "# QWEN3.5-4B UNCENSORED - GOOGLE COLAB T4\n",
39
- "#\n",
40
- "# Model:\n",
41
- "# DreamFast/Qwen3.5-4B-Uncensored-HauhauCS-Aggressive-Safetensor-Benchmark\n",
42
- "#\n",
43
- "# GPU: NVIDIA T4 16GB\n",
44
- "# Quantization: BitsAndBytes NF4 4-bit\n",
45
- "# Compute: FP16\n",
46
- "# Attention: SDPA\n",
47
- "#\n",
48
- "# Features:\n",
49
- "# - Text and image conversations\n",
50
- "# - Conversation history\n",
51
- "# - Separate THINKS and SAYS sections\n",
52
- "# - Larger Gradio chat interface\n",
53
- "# - Expandable latest-response panels\n",
54
- "# - Adjustable generation settings\n",
55
- "# - HF_TOKEN from Colab Secrets\n",
56
- "# ============================================================\n",
57
- "\n",
58
- "\n",
59
- "# ============================================================\n",
60
- "# 1. INSTALL DEPENDENCIES\n",
61
- "# ============================================================\n",
62
- "\n",
63
- "import sys\n",
64
- "import subprocess\n",
65
- "\n",
66
- "subprocess.run(\n",
67
- " [\n",
68
- " sys.executable, \"-m\", \"pip\", \"install\", \"-q\", \"-U\",\n",
69
- " \"bitsandbytes\",\n",
70
- " \"accelerate\",\n",
71
- " \"transformers\",\n",
72
- " \"gradio>=6.0\",\n",
73
- " \"huggingface_hub\",\n",
74
- " \"pillow\"\n",
75
- " ],\n",
76
- " check=True\n",
77
- ")\n",
78
- "\n",
79
- "\n",
80
- "# ============================================================\n",
81
- "# 2. IMPORTS\n",
82
- "# ============================================================\n",
83
- "\n",
84
- "import os\n",
85
- "import re\n",
86
- "import gc\n",
87
- "import torch\n",
88
- "import gradio as gr\n",
89
- "\n",
90
- "from PIL import Image\n",
91
- "from pathlib import Path\n",
92
- "\n",
93
- "from google.colab import userdata\n",
94
- "\n",
95
- "from huggingface_hub import hf_hub_download\n",
96
- "\n",
97
- "from transformers import (\n",
98
- " AutoProcessor,\n",
99
- " AutoTokenizer,\n",
100
- " AutoModelForMultimodalLM,\n",
101
- " BitsAndBytesConfig\n",
102
- ")\n",
103
- "\n",
104
- "\n",
105
- "# ============================================================\n",
106
- "# 3. CONFIGURATION\n",
107
- "# ============================================================\n",
108
- "\n",
109
- "MODEL_ID = (\n",
110
- " \"DreamFast/\"\n",
111
- " \"Qwen3.5-4B-Uncensored-HauhauCS-Aggressive-Safetensor-Benchmark\"\n",
112
- ")\n",
113
- "\n",
114
- "BASE_MODEL_ID = \"Qwen/Qwen3.5-4B\"\n",
115
- "\n",
116
- "HF_TOKEN = userdata.get(\"HF_TOKEN\")\n",
117
- "\n",
118
- "if not HF_TOKEN:\n",
119
- " raise ValueError(\n",
120
- " \"HF_TOKEN not found in Colab Secrets. \"\n",
121
- " \"Please add your Hugging Face token.\"\n",
122
- " )\n",
123
- "\n",
124
- "if not torch.cuda.is_available():\n",
125
- " raise RuntimeError(\"No CUDA GPU detected.\")\n",
126
- "\n",
127
- "print(\"GPU:\", torch.cuda.get_device_name(0))\n",
128
- "print(\"VRAM:\", round(\n",
129
- " torch.cuda.get_device_properties(0).total_memory / 1024**3, 2\n",
130
- "), \"GB\")\n",
131
- "\n",
132
- "torch.backends.cuda.matmul.allow_tf32 = True\n",
133
- "\n",
134
- "\n",
135
- "# ============================================================\n",
136
- "# 4. LOAD CHAT TEMPLATE\n",
137
- "# ============================================================\n",
138
- "\n",
139
- "print(\"\\nLoading Qwen3.5 tokenizer...\")\n",
140
- "\n",
141
- "tokenizer = AutoTokenizer.from_pretrained(\n",
142
- " BASE_MODEL_ID,\n",
143
- " token=HF_TOKEN,\n",
144
- " trust_remote_code=True\n",
145
- ")\n",
146
- "\n",
147
- "if not tokenizer.chat_template:\n",
148
- "\n",
149
- " print(\"Downloading original Qwen3.5 chat template...\")\n",
150
- "\n",
151
- " template_path = hf_hub_download(\n",
152
- " repo_id=BASE_MODEL_ID,\n",
153
- " filename=\"chat_template.jinja\",\n",
154
- " token=HF_TOKEN\n",
155
- " )\n",
156
- "\n",
157
- " with open(template_path, \"r\", encoding=\"utf-8\") as f:\n",
158
- " chat_template = f.read()\n",
159
- "\n",
160
- " tokenizer.chat_template = chat_template\n",
161
- "\n",
162
- "else:\n",
163
- " chat_template = tokenizer.chat_template\n",
164
- "\n",
165
- "print(\"Chat template loaded.\")\n",
166
- "\n",
167
- "\n",
168
- "# ============================================================\n",
169
- "# 5. LOAD PROCESSOR\n",
170
- "# ============================================================\n",
171
- "\n",
172
- "print(\"\\nLoading multimodal processor...\")\n",
173
- "\n",
174
- "processor = AutoProcessor.from_pretrained(\n",
175
- " MODEL_ID,\n",
176
- " token=HF_TOKEN,\n",
177
- " trust_remote_code=True\n",
178
- ")\n",
179
- "\n",
180
- "# Use the compatible original Qwen3.5 tokenizer/template.\n",
181
- "processor.tokenizer = tokenizer\n",
182
- "processor.chat_template = chat_template\n",
183
- "\n",
184
- "print(\"Processor loaded.\")\n",
185
- "\n",
186
- "\n",
187
- "# ============================================================\n",
188
- "# 6. 4-BIT QUANTIZATION\n",
189
- "# ============================================================\n",
190
- "\n",
191
- "quantization_config = BitsAndBytesConfig(\n",
192
- " load_in_4bit=True,\n",
193
- " bnb_4bit_quant_type=\"nf4\",\n",
194
- " bnb_4bit_compute_dtype=torch.float16,\n",
195
- " bnb_4bit_use_double_quant=True\n",
196
- ")\n",
197
- "\n",
198
- "\n",
199
- "# ============================================================\n",
200
- "# 7. LOAD MODEL\n",
201
- "# ============================================================\n",
202
- "\n",
203
- "print(\"\\nLoading model in 4-bit...\")\n",
204
- "\n",
205
- "model = AutoModelForMultimodalLM.from_pretrained(\n",
206
- " MODEL_ID,\n",
207
- " token=HF_TOKEN,\n",
208
- " quantization_config=quantization_config,\n",
209
- " dtype=torch.float16,\n",
210
- " device_map=\"auto\",\n",
211
- " low_cpu_mem_usage=True,\n",
212
- " attn_implementation=\"sdpa\",\n",
213
- " trust_remote_code=True\n",
214
- ")\n",
215
- "\n",
216
- "model.eval()\n",
217
- "\n",
218
- "print(\"\\nModel loaded successfully.\")\n",
219
- "\n",
220
- "print(\"Model device:\", model.get_input_embeddings().weight.device)\n",
221
- "\n",
222
- "# ============================================================\n",
223
- "# 8. PREPARE MULTIMODAL INPUTS\n",
224
- "# ============================================================\n",
225
- "\n",
226
- "def prepare_multimodal_inputs(messages):\n",
227
- "\n",
228
- " # Apply the chat template using the tokenizer directly.\n",
229
- " prompt = tokenizer.apply_chat_template(\n",
230
- " messages,\n",
231
- " tokenize=False,\n",
232
- " add_generation_prompt=True\n",
233
- " )\n",
234
- "\n",
235
- " # Collect images in conversation order.\n",
236
- " images = []\n",
237
- "\n",
238
- " for msg in messages:\n",
239
- "\n",
240
- " content = msg.get(\"content\", [])\n",
241
- "\n",
242
- " if isinstance(content, str):\n",
243
- " continue\n",
244
- "\n",
245
- " if not isinstance(content, list):\n",
246
- " continue\n",
247
- "\n",
248
- " for item in content:\n",
249
- "\n",
250
- " if (\n",
251
- " isinstance(item, dict)\n",
252
- " and item.get(\"type\") == \"image\"\n",
253
- " ):\n",
254
- "\n",
255
- " image = item.get(\"image\")\n",
256
- "\n",
257
- " if image is not None:\n",
258
- " images.append(image)\n",
259
- "\n",
260
- " # Process text and images.\n",
261
- " if images:\n",
262
- "\n",
263
- " formatted_inputs = processor(\n",
264
- " text=prompt,\n",
265
- " images=images,\n",
266
- " return_tensors=\"pt\"\n",
267
- " )\n",
268
- "\n",
269
- " else:\n",
270
- "\n",
271
- " formatted_inputs = processor(\n",
272
- " text=prompt,\n",
273
- " return_tensors=\"pt\"\n",
274
- " )\n",
275
- "\n",
276
- " return formatted_inputs\n",
277
- "\n",
278
- "\n",
279
- "# ============================================================\n",
280
- "# 9. SPLIT THINKING AND FINAL ANSWER\n",
281
- "# ============================================================\n",
282
- "\n",
283
- "def split_thinking_and_answer(raw_text):\n",
284
- "\n",
285
- " \"\"\"\n",
286
- " Separates model-generated <think>...</think> content\n",
287
- " from the final answer.\n",
288
- "\n",
289
- " If the model does not return a think block, its complete\n",
290
- " response is treated as the final answer.\n",
291
- " \"\"\"\n",
292
- "\n",
293
- " raw_text = raw_text.strip()\n",
294
- "\n",
295
- " # Remove common trailing generation tokens.\n",
296
- " raw_text = re.sub(\n",
297
- " r\"<\\|im_end\\|>|<\\|endoftext\\|>|</s>\",\n",
298
- " \"\",\n",
299
- " raw_text\n",
300
- " ).strip()\n",
301
- "\n",
302
- " # Find a complete think block.\n",
303
- " match = re.search(\n",
304
- " r\"<think>(.*?)</think>\",\n",
305
- " raw_text,\n",
306
- " flags=re.DOTALL | re.IGNORECASE\n",
307
- " )\n",
308
- "\n",
309
- " if match:\n",
310
- "\n",
311
- " thinking = match.group(1).strip()\n",
312
- "\n",
313
- " # Everything outside the think block is the answer.\n",
314
- " answer = (\n",
315
- " raw_text[:match.start()]\n",
316
- " + raw_text[match.end():]\n",
317
- " ).strip()\n",
318
- "\n",
319
- " else:\n",
320
- "\n",
321
- " # Handle an unclosed think block.\n",
322
- " open_match = re.search(\n",
323
- " r\"<think>\",\n",
324
- " raw_text,\n",
325
- " flags=re.IGNORECASE\n",
326
- " )\n",
327
- "\n",
328
- " if open_match:\n",
329
- "\n",
330
- " thinking = raw_text[open_match.end():].strip()\n",
331
- " answer = \"\"\n",
332
- "\n",
333
- " else:\n",
334
- "\n",
335
- " thinking = \"The model did not return a separate <think> block.\"\n",
336
- " answer = raw_text\n",
337
- "\n",
338
- " # Remove any remaining template artifacts.\n",
339
- " answer = re.sub(\n",
340
- " r\"<\\|im_start\\|>|<\\|im_sep\\|>|<\\|im_end\\|>\",\n",
341
- " \"\",\n",
342
- " answer\n",
343
- " ).strip()\n",
344
- "\n",
345
- " thinking = re.sub(\n",
346
- " r\"<\\|im_start\\|>|<\\|im_sep\\|>|<\\|im_end\\|>\",\n",
347
- " \"\",\n",
348
- " thinking\n",
349
- " ).strip()\n",
350
- "\n",
351
- " if not answer:\n",
352
- " answer = \"(The model did not produce a separate final answer.)\"\n",
353
- "\n",
354
- " if not thinking:\n",
355
- " thinking = \"(No thinking text was returned.)\"\n",
356
- "\n",
357
- " return thinking, answer\n",
358
- "\n",
359
- "\n",
360
- "# ============================================================\n",
361
- "# 10. GENERATION FUNCTION\n",
362
- "# ============================================================\n",
363
- "\n",
364
- "def generate_response(\n",
365
- " messages,\n",
366
- " max_new_tokens=1024,\n",
367
- " temperature=0.7,\n",
368
- " top_p=0.9\n",
369
- "):\n",
370
- "\n",
371
- " inputs = prepare_multimodal_inputs(messages)\n",
372
- "\n",
373
- " # Move inputs to the model's embedding device.\n",
374
- " input_device = model.get_input_embeddings().weight.device\n",
375
- "\n",
376
- " inputs = inputs.to(input_device)\n",
377
- "\n",
378
- " input_length = inputs[\"input_ids\"].shape[-1]\n",
379
- "\n",
380
- " do_sample = float(temperature) > 0\n",
381
- "\n",
382
- " generation_kwargs = {\n",
383
- " \"max_new_tokens\": int(max_new_tokens),\n",
384
- " \"do_sample\": do_sample,\n",
385
- " \"repetition_penalty\": 1.05,\n",
386
- " \"pad_token_id\": tokenizer.pad_token_id\n",
387
- " if tokenizer.pad_token_id is not None\n",
388
- " else tokenizer.eos_token_id,\n",
389
- " \"eos_token_id\": tokenizer.eos_token_id,\n",
390
- " \"use_cache\": True\n",
391
- " }\n",
392
- "\n",
393
- " if do_sample:\n",
394
- "\n",
395
- " generation_kwargs.update({\n",
396
- " \"temperature\": float(temperature),\n",
397
- " \"top_p\": float(top_p)\n",
398
- " })\n",
399
- "\n",
400
- " with torch.inference_mode():\n",
401
- "\n",
402
- " output_ids = model.generate(\n",
403
- " **inputs,\n",
404
- " **generation_kwargs\n",
405
- " )\n",
406
- "\n",
407
- " # Decode only newly generated tokens.\n",
408
- " generated_ids = output_ids[0][input_length:]\n",
409
- "\n",
410
- " raw_response = tokenizer.decode(\n",
411
- " generated_ids,\n",
412
- " skip_special_tokens=False,\n",
413
- " clean_up_tokenization_spaces=False\n",
414
- " )\n",
415
- "\n",
416
- " thinking, answer = split_thinking_and_answer(raw_response)\n",
417
- "\n",
418
- " # Keep the original model response in the conversation history.\n",
419
- " return raw_response.strip(), thinking, answer\n",
420
- "\n",
421
- "\n",
422
- "# ============================================================\n",
423
- "# 11. GRADIO CHAT FUNCTION\n",
424
- "# ============================================================\n",
425
- "\n",
426
- "def chat(\n",
427
- " message,\n",
428
- " image,\n",
429
- " chat_history,\n",
430
- " conversation_history,\n",
431
- " max_new_tokens,\n",
432
- " temperature,\n",
433
- " top_p\n",
434
- "):\n",
435
- "\n",
436
- " if chat_history is None:\n",
437
- " chat_history = []\n",
438
- "\n",
439
- " if conversation_history is None:\n",
440
- " conversation_history = []\n",
441
- "\n",
442
- " message = (message or \"\").strip()\n",
443
- "\n",
444
- " if not message and image is None:\n",
445
- "\n",
446
- " return (\n",
447
- " chat_history,\n",
448
- " conversation_history,\n",
449
- " \"\",\n",
450
- " None,\n",
451
- " \"\",\n",
452
- " \"Please enter a message or upload an image.\"\n",
453
- " )\n",
454
- "\n",
455
- " # Build user content.\n",
456
- " user_content = []\n",
457
- "\n",
458
- " if message:\n",
459
- "\n",
460
- " user_content.append({\n",
461
- " \"type\": \"text\",\n",
462
- " \"text\": message\n",
463
- " })\n",
464
- "\n",
465
- " if image is not None:\n",
466
- "\n",
467
- " # Convert to RGB PIL image.\n",
468
- " if isinstance(image, Image.Image):\n",
469
- " image = image.convert(\"RGB\")\n",
470
- "\n",
471
- " user_content.append({\n",
472
- " \"type\": \"image\",\n",
473
- " \"image\": image\n",
474
- " })\n",
475
- "\n",
476
- " # Add user turn to actual model conversation.\n",
477
- " conversation_history = list(conversation_history)\n",
478
- "\n",
479
- " conversation_history.append({\n",
480
- " \"role\": \"user\",\n",
481
- " \"content\": user_content\n",
482
- " })\n",
483
- "\n",
484
- " # Display user message in the chat window.\n",
485
- " display_message = message if message else \"Please describe this image.\"\n",
486
- "\n",
487
- " if image is not None:\n",
488
- " display_message += \"\\n\\n📷 *Image attached*\"\n",
489
- "\n",
490
- " chat_history = list(chat_history)\n",
491
- "\n",
492
- " chat_history.append({\n",
493
- " \"role\": \"user\",\n",
494
- " \"content\": display_message\n",
495
- " })\n",
496
- "\n",
497
- " try:\n",
498
- "\n",
499
- " raw_response, thinking, answer = generate_response(\n",
500
- " conversation_history,\n",
501
- " max_new_tokens=max_new_tokens,\n",
502
- " temperature=temperature,\n",
503
- " top_p=top_p\n",
504
- " )\n",
505
- "\n",
506
- " # Store the complete response for subsequent turns.\n",
507
- " conversation_history.append({\n",
508
- " \"role\": \"assistant\",\n",
509
- " \"content\": raw_response\n",
510
- " })\n",
511
- "\n",
512
- " # Add two separate assistant messages to the chat.\n",
513
- " chat_history.append({\n",
514
- " \"role\": \"assistant\",\n",
515
- " \"content\": \"### 💭 WHAT THE MODEL THINKS\\n\\n\" + thinking\n",
516
- " })\n",
517
- "\n",
518
- " chat_history.append({\n",
519
- " \"role\": \"assistant\",\n",
520
- " \"content\": \"### 🗣️ WHAT THE MODEL SAYS\\n\\n\" + answer\n",
521
- " })\n",
522
- "\n",
523
- " return (\n",
524
- " chat_history,\n",
525
- " conversation_history,\n",
526
- " \"\",\n",
527
- " None,\n",
528
- " thinking,\n",
529
- " answer\n",
530
- " )\n",
531
- "\n",
532
- " except Exception as e:\n",
533
- "\n",
534
- " error_message = f\"Generation error:\\n\\n{str(e)}\"\n",
535
- "\n",
536
- " chat_history.append({\n",
537
- " \"role\": \"assistant\",\n",
538
- " \"content\": error_message\n",
539
- " })\n",
540
- "\n",
541
- " return (\n",
542
- " chat_history,\n",
543
- " conversation_history,\n",
544
- " message,\n",
545
- " image,\n",
546
- " \"\",\n",
547
- " error_message\n",
548
- " )\n",
549
- "\n",
550
- "\n",
551
- "# ============================================================\n",
552
- "# 12. CLEAR CONVERSATION\n",
553
- "# ============================================================\n",
554
- "\n",
555
- "def clear_chat():\n",
556
- "\n",
557
- " return (\n",
558
- " [],\n",
559
- " [],\n",
560
- " \"\",\n",
561
- " None,\n",
562
- " \"\",\n",
563
- " \"\"\n",
564
- " )\n",
565
- "\n",
566
- "\n",
567
- "# ============================================================\n",
568
- "# 13. GRADIO INTERFACE\n",
569
- "# ============================================================\n",
570
- "\n",
571
- "CUSTOM_CSS = \"\"\"\n",
572
- "body {\n",
573
- " background: #181818 !important;\n",
574
- "}\n",
575
- "\n",
576
- ".gradio-container {\n",
577
- " max-width: 1500px !important;\n",
578
- " margin: auto !important;\n",
579
- "}\n",
580
- "\n",
581
- "#main-title {\n",
582
- " text-align: center;\n",
583
- " margin-bottom: 5px;\n",
584
- "}\n",
585
- "\n",
586
- "#chat-window {\n",
587
- " min-height: 650px !important;\n",
588
- "}\n",
589
- "\n",
590
- "#chat-window .message {\n",
591
- " font-size: 16px !important;\n",
592
- " line-height: 1.65 !important;\n",
593
- "}\n",
594
- "\n",
595
- "#message-input textarea {\n",
596
- " font-size: 16px !important;\n",
597
- " line-height: 1.6 !important;\n",
598
- "}\n",
599
- "\n",
600
- "#thinking-panel textarea,\n",
601
- "#answer-panel textarea {\n",
602
- " font-size: 15px !important;\n",
603
- " line-height: 1.7 !important;\n",
604
- "}\n",
605
- "\n",
606
- "#send-button {\n",
607
- " background: #a78bfa !important;\n",
608
- " color: #181818 !important;\n",
609
- " font-weight: bold !important;\n",
610
- "}\n",
611
- "\n",
612
- "#clear-button {\n",
613
- " background: #383344 !important;\n",
614
- "}\n",
615
- "\"\"\"\n",
616
- "\n",
617
- "\n",
618
- "with gr.Blocks(\n",
619
- " theme=gr.themes.Base(\n",
620
- " primary_hue=\"violet\",\n",
621
- " secondary_hue=\"orange\",\n",
622
- " neutral_hue=\"slate\"\n",
623
- " ),\n",
624
- " css=CUSTOM_CSS,\n",
625
- " title=\"Qwen3.5-4B Uncensored Chat\"\n",
626
- ") as demo:\n",
627
- "\n",
628
- " gr.Markdown(\n",
629
- " \"\"\"\n",
630
- " # Qwen3.5-4B Uncensored Chat\n",
631
- "\n",
632
- " **DreamFast / HauhauCS — 4-bit NF4 — NVIDIA T4**\n",
633
- "\n",
634
- " Model responses are separated into their generated thinking\n",
635
- " section and their final answer.\n",
636
- " \"\"\",\n",
637
- " elem_id=\"main-title\"\n",
638
- " )\n",
639
- "\n",
640
- " # Actual model conversation and displayed chat are separate.\n",
641
- " conversation_state = gr.State([])\n",
642
- "\n",
643
- " with gr.Row():\n",
644
- "\n",
645
- " with gr.Column(scale=7):\n",
646
- "\n",
647
- " chatbot = gr.Chatbot(\n",
648
- " label=\"Conversation\",\n",
649
- " height=650,\n",
650
- " autoscroll=True,\n",
651
- " elem_id=\"chat-window\"\n",
652
- " )\n",
653
- "\n",
654
- " with gr.Row():\n",
655
- "\n",
656
- " message_input = gr.Textbox(\n",
657
- " label=\"Your message\",\n",
658
- " placeholder=\"Write your message here...\",\n",
659
- " lines=4,\n",
660
- " max_lines=10,\n",
661
- " scale=5,\n",
662
- " elem_id=\"message-input\"\n",
663
- " )\n",
664
- "\n",
665
- " image_input = gr.Image(\n",
666
- " label=\"Optional image\",\n",
667
- " type=\"pil\",\n",
668
- " height=150,\n",
669
- " sources=[\"upload\", \"clipboard\"],\n",
670
- " scale=2\n",
671
- " )\n",
672
- "\n",
673
- " with gr.Row():\n",
674
- "\n",
675
- " send_button = gr.Button(\n",
676
- " \"Send\",\n",
677
- " variant=\"primary\",\n",
678
- " elem_id=\"send-button\",\n",
679
- " scale=3\n",
680
- " )\n",
681
- "\n",
682
- " clear_button = gr.Button(\n",
683
- " \"Clear conversation\",\n",
684
- " variant=\"secondary\",\n",
685
- " elem_id=\"clear-button\",\n",
686
- " scale=1\n",
687
- " )\n",
688
- "\n",
689
- " with gr.Column(scale=3):\n",
690
- "\n",
691
- " gr.Markdown(\"## Generation settings\")\n",
692
- "\n",
693
- " max_new_tokens = gr.Slider(\n",
694
- " minimum=128,\n",
695
- " maximum=4096,\n",
696
- " value=1024,\n",
697
- " step=128,\n",
698
- " label=\"Maximum new tokens\"\n",
699
- " )\n",
700
- "\n",
701
- " temperature = gr.Slider(\n",
702
- " minimum=0.0,\n",
703
- " maximum=1.5,\n",
704
- " value=0.7,\n",
705
- " step=0.05,\n",
706
- " label=\"Temperature\"\n",
707
- " )\n",
708
- "\n",
709
- " top_p = gr.Slider(\n",
710
- " minimum=0.1,\n",
711
- " maximum=1.0,\n",
712
- " value=0.9,\n",
713
- " step=0.05,\n",
714
- " label=\"Top-p\"\n",
715
- " )\n",
716
- "\n",
717
- " gr.Markdown(\n",
718
- " \"\"\"\n",
719
- " **Tip:** Set temperature to `0` for deterministic\n",
720
- " generation. Increase maximum tokens for longer responses.\n",
721
- " \"\"\"\n",
722
- " )\n",
723
- "\n",
724
- " gr.Markdown(\"---\")\n",
725
- "\n",
726
- " gr.Markdown(\"## Latest model response\")\n",
727
- "\n",
728
- " with gr.Row():\n",
729
- "\n",
730
- " with gr.Column():\n",
731
- "\n",
732
- " thinking_output = gr.Textbox(\n",
733
- " label=\"💭 What the model thinks\",\n",
734
- " lines=12,\n",
735
- " max_lines=25,\n",
736
- " interactive=False,\n",
737
- " placeholder=\"Generated thinking will appear here...\",\n",
738
- " elem_id=\"thinking-panel\"\n",
739
- " )\n",
740
- "\n",
741
- " with gr.Column():\n",
742
- "\n",
743
- " answer_output = gr.Textbox(\n",
744
- " label=\"🗣️ What the model says\",\n",
745
- " lines=12,\n",
746
- " max_lines=25,\n",
747
- " interactive=False,\n",
748
- " placeholder=\"The final answer will appear here...\",\n",
749
- " elem_id=\"answer-panel\"\n",
750
- " )\n",
751
- "\n",
752
- " # ========================================================\n",
753
- " # EVENTS\n",
754
- " # ========================================================\n",
755
- "\n",
756
- " inputs = [\n",
757
- " message_input,\n",
758
- " image_input,\n",
759
- " chatbot,\n",
760
- " conversation_state,\n",
761
- " max_new_tokens,\n",
762
- " temperature,\n",
763
- " top_p\n",
764
- " ]\n",
765
- "\n",
766
- " outputs = [\n",
767
- " chatbot,\n",
768
- " conversation_state,\n",
769
- " message_input,\n",
770
- " image_input,\n",
771
- " thinking_output,\n",
772
- " answer_output\n",
773
- " ]\n",
774
- "\n",
775
- " send_button.click(\n",
776
- " fn=chat,\n",
777
- " inputs=inputs,\n",
778
- " outputs=outputs\n",
779
- " )\n",
780
- "\n",
781
- " message_input.submit(\n",
782
- " fn=chat,\n",
783
- " inputs=inputs,\n",
784
- " outputs=outputs\n",
785
- " )\n",
786
- "\n",
787
- " clear_button.click(\n",
788
- " fn=clear_chat,\n",
789
- " inputs=[],\n",
790
- " outputs=outputs\n",
791
- " )\n",
792
- "\n",
793
- "\n",
794
- "# ============================================================\n",
795
- "# 14. LAUNCH\n",
796
- "# ============================================================\n",
797
- "\n",
798
- "print(\"\\nStarting Gradio interface...\")\n",
799
- "\n",
800
- "demo.queue().launch(\n",
801
- " share=True,\n",
802
- " debug=True,\n",
803
- " show_error=True\n",
804
- ")\n"
805
- ]
806
- }
807
- ]
808
- }
 
1
+ {"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[{"file_id":"https://huggingface.co/datasets/codeShare/lora-training-data/blob/main/HauauCS_Abliterated_Qwen_3_5_4B_LLM_model.ipynb","timestamp":1794394717019}],"gpuType":"T4"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU"},"cells":[{"cell_type":"markdown","source":["# HauauCS Abliterated Qwen 3.5 -4B LLM model (image and/or text input)"],"metadata":{"id":"ewVsp7wRAH2j"}},{"cell_type":"code","execution_count":null,"metadata":{"id":"lC55DC9m_4Zz"},"outputs":[],"source":["\n","# ============================================================\n","# QWEN3.5-4B UNCENSORED - GOOGLE COLAB T4\n","#\n","# Model:\n","# DreamFast/Qwen3.5-4B-Uncensored-HauhauCS-Aggressive-Safetensor-Benchmark\n","#\n","# GPU: NVIDIA T4 16GB\n","# Quantization: BitsAndBytes NF4 4-bit\n","# Compute: FP16\n","# Attention: SDPA\n","#\n","# Features:\n","# - Text and image conversations\n","# - Conversation history\n","# - Separate THINKS and SAYS sections\n","# - Larger Gradio chat interface\n","# - Expandable latest-response panels\n","# - Adjustable generation settings\n","# - HF_TOKEN from Colab Secrets\n","# ============================================================\n","\n","\n","# ============================================================\n","# 1. INSTALL DEPENDENCIES\n","# ============================================================\n","\n","import sys\n","import subprocess\n","\n","# Explicitly upgrade Pillow to ensure correct typing dependencies are resolved.\n","subprocess.run(\n"," [\n"," sys.executable, \"-m\", \"pip\", \"install\", \"-q\", \"-U\",\n"," \"--force-reinstall\",\n"," \"pillow\",\n"," \"bitsandbytes\",\n"," \"accelerate\",\n"," \"transformers\",\n"," \"gradio>=6.0\",\n"," \"huggingface_hub\"\n"," ],\n"," check=True\n",")\n","\n","import transformers\n","# ============================================================\n","# 2. IMPORTS\n","# ============================================================\n","\n","import os\n","import re\n","import gc\n","import torch\n","import gradio as gr\n","\n","from PIL import Image\n","from pathlib import Path\n","\n","from google.colab import userdata\n","\n","from huggingface_hub import hf_hub_download\n","\n","from transformers import (\n"," AutoProcessor,\n"," AutoTokenizer,\n"," AutoModelForMultimodalLM,\n"," BitsAndBytesConfig\n",")\n","\n","\n","# ============================================================\n","# 3. CONFIGURATION\n","# ============================================================\n","\n","MODEL_ID = (\n"," \"DreamFast/\"\n"," \"Qwen3.5-4B-Uncensored-HauhauCS-Aggressive-Safetensor-Benchmark\"\n",")\n","\n","BASE_MODEL_ID = \"Qwen/Qwen3.5-4B\"\n","\n","HF_TOKEN = userdata.get(\"HF_TOKEN\")\n","\n","if not HF_TOKEN:\n"," raise ValueError(\n"," \"HF_TOKEN not found in Colab Secrets. \"\n"," \"Please add your Hugging Face token.\"\n"," )\n","\n","if not torch.cuda.is_available():\n"," raise RuntimeError(\"No CUDA GPU detected.\")\n","\n","print(\"GPU:\", torch.cuda.get_device_name(0))\n","print(\"VRAM:\", round(\n"," torch.cuda.get_device_properties(0).total_memory / 1024**3, 2\n","), \"GB\")\n","\n","torch.backends.cuda.matmul.allow_tf32 = True\n","\n","\n","# ============================================================\n","# 4. LOAD CHAT TEMPLATE\n","# ============================================================\n","\n","print(\"\\nLoading Qwen3.5 tokenizer...\")\n","\n","tokenizer = AutoTokenizer.from_pretrained(\n"," BASE_MODEL_ID,\n"," token=HF_TOKEN,\n"," trust_remote_code=True\n",")\n","\n","if not tokenizer.chat_template:\n","\n"," print(\"Downloading original Qwen3.5 chat template...\")\n","\n"," template_path = hf_hub_download(\n"," repo_id=BASE_MODEL_ID,\n"," filename=\"chat_template.jinja\",\n"," token=HF_TOKEN\n"," )\n","\n"," with open(template_path, \"r\", encoding=\"utf-8\") as f:\n"," chat_template = f.read()\n","\n"," tokenizer.chat_template = chat_template\n","\n","else:\n"," chat_template = tokenizer.chat_template\n","\n","print(\"Chat template loaded.\")\n","\n","\n","# ============================================================\n","# 5. LOAD PROCESSOR\n","# ============================================================\n","\n","print(\"\\nLoading multimodal processor...\")\n","\n","processor = AutoProcessor.from_pretrained(\n"," MODEL_ID,\n"," token=HF_TOKEN,\n"," trust_remote_code=True\n",")\n","\n","# Use the compatible original Qwen3.5 tokenizer/template.\n","processor.tokenizer = tokenizer\n","processor.chat_template = chat_template\n","\n","print(\"Processor loaded.\")\n","\n","\n","# ============================================================\n","# 6. 4-BIT QUANTIZATION\n","# ============================================================\n","\n","quantization_config = BitsAndBytesConfig(\n"," load_in_4bit=True,\n"," bnb_4bit_quant_type=\"nf4\",\n"," bnb_4bit_compute_dtype=torch.float16,\n"," bnb_4bit_use_double_quant=True\n",")\n","\n","\n","# ============================================================\n","# 7. LOAD MODEL\n","# ============================================================\n","\n","print(\"\\nLoading model in 4-bit...\")\n","\n","model = AutoModelForMultimodalLM.from_pretrained(\n"," MODEL_ID,\n"," token=HF_TOKEN,\n"," quantization_config=quantization_config,\n"," dtype=torch.float16,\n"," device_map=\"auto\",\n"," low_cpu_mem_usage=True,\n"," attn_implementation=\"sdpa\",\n"," trust_remote_code=True\n",")\n","\n","model.eval()\n","\n","print(\"\\nModel loaded successfully.\")\n","\n","print(\"Model device:\", model.get_input_embeddings().weight.device)\n","\n","# ============================================================\n","# 8. PREPARE MULTIMODAL INPUTS\n","# ============================================================\n","\n","def prepare_multimodal_inputs(messages):\n","\n"," # Apply the chat template using the tokenizer directly.\n"," prompt = tokenizer.apply_chat_template(\n"," messages,\n"," tokenize=False,\n"," add_generation_prompt=True\n"," )\n","\n"," # Collect images in conversation order.\n"," images = []\n","\n"," for msg in messages:\n","\n"," content = msg.get(\"content\", [])\n","\n"," if isinstance(content, str):\n"," continue\n","\n"," if not isinstance(content, list):\n"," continue\n","\n"," for item in content:\n","\n"," if (\n"," isinstance(item, dict)\n"," and item.get(\"type\") == \"image\"\n"," ):\n","\n"," image = item.get(\"image\")\n","\n"," if image is not None:\n"," images.append(image)\n","\n"," # Process text and images.\n"," if images:\n","\n"," formatted_inputs = processor(\n"," text=prompt,\n"," images=images,\n"," return_tensors=\"pt\"\n"," )\n","\n"," else:\n","\n"," formatted_inputs = processor(\n"," text=prompt,\n"," return_tensors=\"pt\"\n"," )\n","\n"," return formatted_inputs\n","\n","\n","# ============================================================\n","# 9. SPLIT THINKING AND FINAL ANSWER\n","# ============================================================\n","\n","def split_thinking_and_answer(raw_text):\n","\n"," \"\"\"\n"," Separates model-generated <think>...</think> content\n"," from the final answer.\n","\n"," If the model does not return a think block, its complete\n"," response is treated as the final answer.\n"," \"\"\"\n","\n"," raw_text = raw_text.strip()\n","\n"," # Remove common trailing generation tokens.\n"," raw_text = re.sub(\n"," r\"<\\|im_end\\|>|<\\|endoftext\\|>|</s>\",\n"," \"\",\n"," raw_text\n"," ).strip()\n","\n"," # Find a complete think block.\n"," match = re.search(\n"," r\"<think>(.*?)</think>\",\n"," raw_text,\n"," flags=re.DOTALL | re.IGNORECASE\n"," )\n","\n"," if match:\n","\n"," thinking = match.group(1).strip()\n","\n"," # Everything outside the think block is the answer.\n"," answer = (\n"," raw_text[:match.start()]\n"," + raw_text[match.end():]\n"," ).strip()\n","\n"," else:\n","\n"," # Handle an unclosed think block.\n"," open_match = re.search(\n"," r\"<think>\",\n"," raw_text,\n"," flags=re.IGNORECASE\n"," )\n","\n"," if open_match:\n","\n"," thinking = raw_text[open_match.end():].strip()\n"," answer = \"\"\n","\n"," else:\n","\n"," thinking = \"The model did not return a separate <think> block.\"\n"," answer = raw_text\n","\n"," # Remove any remaining template artifacts.\n"," answer = re.sub(\n"," r\"<\\|im_start\\|>|<\\|im_sep\\|>|<\\|im_end\\|>\",\n"," \"\",\n"," answer\n"," ).strip()\n","\n"," thinking = re.sub(\n"," r\"<\\|im_start\\|>|<\\|im_sep\\|>|<\\|im_end\\|>\",\n"," \"\",\n"," thinking\n"," ).strip()\n","\n"," if not answer:\n"," answer = \"(The model did not produce a separate final answer.)\"\n","\n"," if not thinking:\n"," thinking = \"(No thinking text was returned.)\"\n","\n"," return thinking, answer\n","\n","\n","# ============================================================\n","# 10. GENERATION FUNCTION\n","# ============================================================\n","\n","def generate_response(\n"," messages,\n"," max_new_tokens=1024,\n"," temperature=0.7,\n"," top_p=0.9\n","):\n","\n"," inputs = prepare_multimodal_inputs(messages)\n","\n"," # Move inputs to the model's embedding device.\n"," input_device = model.get_input_embeddings().weight.device\n","\n"," inputs = inputs.to(input_device)\n","\n"," input_length = inputs[\"input_ids\"].shape[-1]\n","\n"," do_sample = float(temperature) > 0\n","\n"," generation_kwargs = {\n"," \"max_new_tokens\": int(max_new_tokens),\n"," \"do_sample\": do_sample,\n"," \"repetition_penalty\": 1.05,\n"," \"pad_token_id\": tokenizer.pad_token_id\n"," if tokenizer.pad_token_id is not None\n"," else tokenizer.eos_token_id,\n"," \"eos_token_id\": tokenizer.eos_token_id,\n"," \"use_cache\": True\n"," }\n","\n"," if do_sample:\n","\n"," generation_kwargs.update({\n"," \"temperature\": float(temperature),\n"," \"top_p\": float(top_p)\n"," })\n","\n"," with torch.inference_mode():\n","\n"," output_ids = model.generate(\n"," **inputs,\n"," **generation_kwargs\n"," )\n","\n"," # Decode only newly generated tokens.\n"," generated_ids = output_ids[0][input_length:]\n","\n"," raw_response = tokenizer.decode(\n"," generated_ids,\n"," skip_special_tokens=False,\n"," clean_up_tokenization_spaces=False\n"," )\n","\n"," thinking, answer = split_thinking_and_answer(raw_response)\n","\n"," # Keep the original model response in the conversation history.\n"," return raw_response.strip(), thinking, answer\n","\n","\n","# ============================================================\n","# 11. GRADIO CHAT FUNCTION\n","# ============================================================\n","\n","def chat(\n"," message,\n"," image,\n"," chat_history,\n"," conversation_history,\n"," max_new_tokens,\n"," temperature,\n"," top_p\n","):\n","\n"," if chat_history is None:\n"," chat_history = []\n","\n"," if conversation_history is None:\n"," conversation_history = []\n","\n"," message = (message or \"\").strip()\n","\n"," if not message and image is None:\n","\n"," return (\n"," chat_history,\n"," conversation_history,\n"," \"\",\n"," None,\n"," \"\",\n"," \"Please enter a message or upload an image.\"\n"," )\n","\n"," # Build user content.\n"," user_content = []\n","\n"," if message:\n","\n"," user_content.append({\n"," \"type\": \"text\",\n"," \"text\": message\n"," })\n","\n"," if image is not None:\n","\n"," # Convert to RGB PIL image.\n"," if isinstance(image, Image.Image):\n"," image = image.convert(\"RGB\")\n","\n"," user_content.append({\n"," \"type\": \"image\",\n"," \"image\": image\n"," })\n","\n"," # Add user turn to actual model conversation.\n"," conversation_history = list(conversation_history)\n","\n"," conversation_history.append({\n"," \"role\": \"user\",\n"," \"content\": user_content\n"," })\n","\n"," # Display user message in the chat window.\n"," display_message = message if message else \"Please describe this image.\"\n","\n"," if image is not None:\n"," display_message += \"\\n\\n📷 *Image attached*\"\n","\n"," chat_history = list(chat_history)\n","\n"," chat_history.append({\n"," \"role\": \"user\",\n"," \"content\": display_message\n"," })\n","\n"," try:\n","\n"," raw_response, thinking, answer = generate_response(\n"," conversation_history,\n"," max_new_tokens=max_new_tokens,\n"," temperature=temperature,\n"," top_p=top_p\n"," )\n","\n"," # Store the complete response for subsequent turns.\n"," conversation_history.append({\n"," \"role\": \"assistant\",\n"," \"content\": raw_response\n"," })\n","\n"," # Add two separate assistant messages to the chat.\n"," chat_history.append({\n"," \"role\": \"assistant\",\n"," \"content\": \"### 💭 WHAT THE MODEL THINKS\\n\\n\" + thinking\n"," })\n","\n"," chat_history.append({\n"," \"role\": \"assistant\",\n"," \"content\": \"### 🗣️ WHAT THE MODEL SAYS\\n\\n\" + answer\n"," })\n","\n"," return (\n"," chat_history,\n"," conversation_history,\n"," \"\",\n"," None,\n"," thinking,\n"," answer\n"," )\n","\n"," except Exception as e:\n","\n"," error_message = f\"Generation error:\\n\\n{str(e)}\"\n","\n"," chat_history.append({\n"," \"role\": \"assistant\",\n"," \"content\": error_message\n"," })\n","\n"," return (\n"," chat_history,\n"," conversation_history,\n"," message,\n"," image,\n"," \"\",\n"," error_message\n"," )\n","\n","\n","# ============================================================\n","# 12. CLEAR CONVERSATION\n","# ============================================================\n","\n","def clear_chat():\n","\n"," return (\n"," [],\n"," [],\n"," \"\",\n"," None,\n"," \"\",\n"," \"\"\n"," )\n","\n","\n","# ============================================================\n","# 13. GRADIO INTERFACE\n","# ============================================================\n","\n","CUSTOM_CSS = \"\"\"\n","body {\n"," background: #181818 !important;\n","}\n","\n",".gradio-container {\n"," max-width: 1500px !important;\n"," margin: auto !important;\n","}\n","\n","#main-title {\n"," text-align: center;\n"," margin-bottom: 5px;\n","}\n","\n","#chat-window {\n"," min-height: 650px !important;\n","}\n","\n","#chat-window .message {\n"," font-size: 16px !important;\n"," line-height: 1.65 !important;\n","}\n","\n","#message-input textarea {\n"," font-size: 16px !important;\n"," line-height: 1.6 !important;\n","}\n","\n","#thinking-panel textarea,\n","#answer-panel textarea {\n"," font-size: 15px !important;\n"," line-height: 1.7 !important;\n","}\n","\n","#send-button {\n"," background: #a78bfa !important;\n"," color: #181818 !important;\n"," font-weight: bold !important;\n","}\n","\n","#clear-button {\n"," background: #383344 !important;\n","}\n","\"\"\"\n","\n","\n","with gr.Blocks(\n"," theme=gr.themes.Base(\n"," primary_hue=\"violet\",\n"," secondary_hue=\"orange\",\n"," neutral_hue=\"slate\"\n"," ),\n"," css=CUSTOM_CSS,\n"," title=\"Qwen3.5-4B Uncensored Chat\"\n",") as demo:\n","\n"," gr.Markdown(\n"," \"\"\"\n"," # Qwen3.5-4B Uncensored Chat\n","\n"," **DreamFast / HauhauCS — 4-bit NF4 — NVIDIA T4**\n","\n"," Model responses are separated into their generated thinking\n"," section and their final answer.\n"," \"\"\",\n"," elem_id=\"main-title\"\n"," )\n","\n"," # Actual model conversation and displayed chat are separate.\n"," conversation_state = gr.State([])\n","\n"," with gr.Row():\n","\n"," with gr.Column(scale=7):\n","\n"," chatbot = gr.Chatbot(\n"," label=\"Conversation\",\n"," height=650,\n"," autoscroll=True,\n"," elem_id=\"chat-window\"\n"," )\n","\n"," with gr.Row():\n","\n"," message_input = gr.Textbox(\n"," label=\"Your message\",\n"," placeholder=\"Write your message here...\",\n"," lines=4,\n"," max_lines=10,\n"," scale=5,\n"," elem_id=\"message-input\"\n"," )\n","\n"," image_input = gr.Image(\n"," label=\"Optional image\",\n"," type=\"pil\",\n"," height=150,\n"," sources=[\"upload\", \"clipboard\"],\n"," scale=2\n"," )\n","\n"," with gr.Row():\n","\n"," send_button = gr.Button(\n"," \"Send\",\n"," variant=\"primary\",\n"," elem_id=\"send-button\",\n"," scale=3\n"," )\n","\n"," clear_button = gr.Button(\n"," \"Clear conversation\",\n"," variant=\"secondary\",\n"," elem_id=\"clear-button\",\n"," scale=1\n"," )\n","\n"," with gr.Column(scale=3):\n","\n"," gr.Markdown(\"## Generation settings\")\n","\n"," max_new_tokens = gr.Slider(\n"," minimum=128,\n"," maximum=4096,\n"," value=1024,\n"," step=128,\n"," label=\"Maximum new tokens\"\n"," )\n","\n"," temperature = gr.Slider(\n"," minimum=0.0,\n"," maximum=1.5,\n"," value=0.7,\n"," step=0.05,\n"," label=\"Temperature\"\n"," )\n","\n"," top_p = gr.Slider(\n"," minimum=0.1,\n"," maximum=1.0,\n"," value=0.9,\n"," step=0.05,\n"," label=\"Top-p\"\n"," )\n","\n"," gr.Markdown(\n"," \"\"\"\n"," **Tip:** Set temperature to `0` for deterministic\n"," generation. Increase maximum tokens for longer responses.\n"," \"\"\"\n"," )\n","\n"," gr.Markdown(\"---\")\n","\n"," gr.Markdown(\"## Latest model response\")\n","\n"," with gr.Row():\n","\n"," with gr.Column():\n","\n"," thinking_output = gr.Textbox(\n"," label=\"💭 What the model thinks\",\n"," lines=12,\n"," max_lines=25,\n"," interactive=False,\n"," placeholder=\"Generated thinking will appear here...\",\n"," elem_id=\"thinking-panel\"\n"," )\n","\n"," with gr.Column():\n","\n"," answer_output = gr.Textbox(\n"," label=\"🗣️ What the model says\",\n"," lines=12,\n"," max_lines=25,\n"," interactive=False,\n"," placeholder=\"The final answer will appear here...\",\n"," elem_id=\"answer-panel\"\n"," )\n","\n"," # ========================================================\n"," # EVENTS\n"," # ========================================================\n","\n"," inputs = [\n"," message_input,\n"," image_input,\n"," chatbot,\n"," conversation_state,\n"," max_new_tokens,\n"," temperature,\n"," top_p\n"," ]\n","\n"," outputs = [\n"," chatbot,\n"," conversation_state,\n"," message_input,\n"," image_input,\n"," thinking_output,\n"," answer_output\n"," ]\n","\n"," send_button.click(\n"," fn=chat,\n"," inputs=inputs,\n"," outputs=outputs\n"," )\n","\n"," message_input.submit(\n"," fn=chat,\n"," inputs=inputs,\n"," outputs=outputs\n"," )\n","\n"," clear_button.click(\n"," fn=clear_chat,\n"," inputs=[],\n"," outputs=outputs\n"," )\n","\n","\n","# ============================================================\n","# 14. LAUNCH\n","# ============================================================\n","\n","print(\"\\nStarting Gradio interface...\")\n","\n","demo.queue().launch(\n"," share=True,\n"," debug=True,\n"," show_error=True\n",")\n"]}]}