Hugging Face's logo

ZAQERIS
/
ZAQION

Image-Text-to-Text
Transformers
Safetensors
PyTorch
English
Arabic
multilingual
zaqeris_zaqion
text-generation
multimodal
vision
150b
moe-150b
150b-moe
mixture-of-experts
48-experts
150-billion-parameters
large-model
large-language-model
causal-lm
vllm
code
math
reasoning
arabic
egypt
zaqeris
m4-gold
lattice-quantization
conversational
custom_code
Eval Results (legacy)
zaqeris_m4_gold
Model card Files
xet
 Community
1
ZAQION
File size: 115,832 Bytes
8859c2e
 
 
 
 
bd92f58
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bd92f58
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bd92f58
8859c2e
 
bd92f58
8859c2e
 
2b3d559
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
98d517b
 
 
8859c2e
 
 
 
 
2b3d559
8859c2e
2b3d559
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
2b3d559
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bd92f58
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f730fe8
 
8859c2e
 
 
f730fe8
 
8859c2e
 
f730fe8
 
 
8859c2e
 
 
 
 
 
 
 
37df5ae
 
8859c2e
 
 
98d517b
 
 
 
 
 
 
 
 
bd92f58
98d517b
 
 
 
 
 
 
f730fe8
98d517b
 
 
 
 
 
f730fe8
98d517b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bd92f58
98d517b
 
 
 
 
 
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bd92f58
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bd92f58
8859c2e
 
bd92f58
8859c2e
 
 
 
 
 
 
 
 
 
 
 
bd92f58
8859c2e
 
bd92f58
 
2b3d559
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
2b3d559
 
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
98d517b
 
 
bd92f58
98d517b
37df5ae
98d517b
f730fe8
98d517b
 
37df5ae
98d517b
 
bd92f58
8859c2e
 
 
bd92f58
8859c2e
 
bd92f58
 
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
37df5ae
2b3d559
 
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37df5ae
98d517b
 
 
37df5ae
 
 
98d517b
37df5ae
98d517b
 
37df5ae
98d517b
 
bd92f58
8859c2e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
<!DOCTYPE html>
<html lang="en" class="dark">
<head>
  <meta charset="UTF-8">
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  <title>ZAQERIS AI — Independent Foundation Architecture Explorer & Empirical Dossier</title>
  <script src="https://cdn.tailwindcss.com"></script>
  <script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
  <script>

    tailwind.config = {

      darkMode: 'class',

      theme: {

        extend: {

          colors: {

            academic: {

              950: '#050608',

              900: '#0A0C10',

              850: '#0F1218',

              800: '#161B24',

              750: '#1D2432',

              700: '#252E3E',

              600: '#344054',

              500: '#475467',

              400: '#98A2B3',

              300: '#D0D5DD',

              200: '#E4E7EC',

              100: '#F8F9FC',

            },

            zaqeris: {

              gold: '#D4AF37',

              darkgold: '#997A15',

              bronze: '#8C7853',

              slate: '#334155',

              blue: '#1E3A8A',

            }

          }

        }

      }

    }

  </script>
  <style>

    @import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700&family=JetBrains+Mono:wght@400;500;600&family=Cairo:wght@400;600;700;800&display=swap');

    body { font-family: 'Inter', -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, sans-serif; }

    body.lang-ar { font-family: 'Cairo', -apple-system, BlinkMacSystemFont, sans-serif; direction: rtl; }

    code, pre, .font-mono { font-family: 'JetBrains Mono', monospace; }

    .tab-active {

      border-bottom-color: #D4AF37 !important;

      color: #F8F9FC !important;

      background-color: #161B24 !important;

    }

  </style>
</head>
<body class="bg-academic-950 text-academic-300 min-h-screen flex flex-col antialiased">

  <!-- TOP HEADER -->
  <header class="border-b border-academic-750 bg-academic-900 sticky top-0 z-50 shadow-md">
    <div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 h-20 flex items-center justify-between">
      <div class="flex items-center space-x-4 space-x-reverse">
        <img src="data:image/jpeg;base64,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" alt="ZAQERIS AI Logo" class="w-12 h-12 rounded-lg border border-academic-700 object-cover shadow-sm">
        <div>
          <div class="flex items-center space-x-2 space-x-reverse">
            <span class="font-bold text-base text-academic-100 tracking-wider uppercase">ZAQERIS AI</span>
            <span class="text-[10px] font-semibold px-2 py-0.5 rounded bg-academic-800 text-zaqeris-gold border border-academic-700 tracking-widest uppercase" data-i18n="badge_lab">RESEARCH LAB</span>
            <span class="text-[10px] font-mono px-2 py-0.5 rounded bg-academic-800 text-academic-400 border border-academic-750">EGYPT</span>
          </div>
          <p class="text-xs text-academic-400 mt-0.5" data-i18n="sub_header">Independent AI Research Group • Foundation Model Directorate</p>
        </div>
      </div>
      
      <div class="flex items-center space-x-3 space-x-reverse">
        <!-- LANGUAGE TOGGLE -->
        <button onclick="toggleLanguage()" id="lang-toggle-btn" class="inline-flex items-center space-x-1.5 space-x-reverse text-xs font-semibold bg-academic-850 hover:bg-academic-800 text-academic-200 border border-academic-700 px-3.5 py-2 rounded transition-all shadow-sm">
          <svg class="w-3.5 h-3.5 text-zaqeris-gold" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M3 5h12M9 3v2m1.048 9.5A18.022 18.022 0 016.412 9m6.088 9h7M11 21l5-10 5 10M12.751 5C11.783 10.77 8.07 15.61 3 18.129"></path></svg>
          <span id="lang-label">العربية</span>
        </button>

        <a href="https://huggingface.co/ZAQERIS/ZAQION" target="_blank" class="inline-flex items-center space-x-2 space-x-reverse text-xs font-medium bg-academic-800 hover:bg-academic-750 text-academic-100 border border-academic-650 px-4 py-2 rounded transition-colors shadow-sm">
          <span data-i18n="hf_repo_btn">Hugging Face Hub</span>
          <svg class="w-3.5 h-3.5 text-academic-400" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M10 6H6a2 2 0 00-2 2v10a2 2 0 002 2h10a2 2 0 002-2v-4M14 4h6m0 0v6m0-6L10 14"></path></svg>
        </a>
      </div>
    </div>
  </header>

  <!-- HERO SECTION -->
  <section class="border-b border-academic-750 bg-academic-900 py-10">
    <div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8">
      <div class="max-w-4xl">
        <div class="inline-flex items-center space-x-2 space-x-reverse text-xs font-mono text-zaqeris-gold mb-3 uppercase tracking-wider">
          <span data-i18n="hero_pill_1">Official Technical Architecture Explorer</span>
          <span>•</span>
          <span data-i18n="hero_pill_2">Foundation Suite</span>
        </div>
        <h1 class="text-3xl sm:text-4xl font-bold text-academic-100 tracking-tight leading-tight" data-i18n="hero_title">
          ZAQION 1.0 (150B MoE) & Foundation Model Architecture Suite
        </h1>
        <p class="text-sm text-academic-400 mt-3 leading-relaxed" data-i18n="hero_desc">
          Official engineering specifications, mathematical lattice quantization benchmarks, and comparative architectural analysis of the AXIOM (Lightweight) and ZAQION (Heavyweight Large Model) foundation model families.
        </p>

        <!-- STATS BADGES -->
        <div class="grid grid-cols-2 sm:grid-cols-4 gap-3 mt-8">
          <div class="border border-academic-750 bg-academic-950 p-3.5 rounded shadow-sm">
            <div class="text-[11px] font-mono uppercase text-academic-500" data-i18n="stat_1_label">ZAQION Physical Capacity</div>
            <div class="text-lg font-bold text-academic-100 mt-1">150,000,000,000</div>
            <div class="text-[11px] text-zaqeris-gold" data-i18n="stat_1_sub">50.0B Peak Active (Adaptive-k)</div>
          </div>
          <div class="border border-academic-750 bg-academic-950 p-3.5 rounded shadow-sm">
            <div class="text-[11px] font-mono uppercase text-academic-500" data-i18n="stat_2_label">AXIOM Physical Capacity</div>
            <div class="text-lg font-bold text-academic-100 mt-1">79,425,295,488</div>
            <div class="text-[11px] text-academic-400" data-i18n="stat_2_sub">12.39B Active (Lightweight Edge)</div>
          </div>
          <div class="border border-academic-750 bg-academic-950 p-3.5 rounded shadow-sm">
            <div class="text-[11px] font-mono uppercase text-academic-500" data-i18n="stat_3_label">ZAQERIS-M4 Footprint</div>
            <div class="text-lg font-bold text-academic-100 mt-1">48.9 GiB (2.115 bpp)</div>
            <div class="text-[11px] text-zaqeris-gold" data-i18n="stat_3_sub">16D Barnes-Wall Lattice</div>
          </div>
          <div class="border border-academic-750 bg-academic-950 p-3.5 rounded shadow-sm">
            <div class="text-[11px] font-mono uppercase text-academic-500" data-i18n="stat_4_label">From-Scratch Pre-Training</div>
            <div class="text-lg font-bold text-academic-100 mt-1">3.2 Trillion Tokens</div>
            <div class="text-[11px] text-academic-400" data-i18n="stat_4_sub">512× H100 InfiniBand Cluster</div>
          </div>
        </div>
      </div>
    </div>
  </section>

  <!-- NAVIGATION TABS -->
  <main class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 py-8 flex-1 w-full">
    <div class="flex border-b border-academic-750 space-x-2 space-x-reverse mb-8 overflow-x-auto pb-1">
      <button onclick="switchTab('family')" id="tab-btn-family" class="tab-btn tab-active px-4 py-2.5 text-xs font-semibold uppercase tracking-wider rounded-t border-b-2 transition-all">
        <span data-i18n="tab_1">1. AXIOM vs ZAQION Family</span>
      </button>
      <button onclick="switchTab('lattice')" id="tab-btn-lattice" class="tab-btn border-transparent text-academic-400 hover:text-academic-200 px-4 py-2.5 text-xs font-semibold uppercase tracking-wider rounded-t border-b-2 transition-all">
        <span data-i18n="tab_2">2. ZAQERIS-M4 vs M3</span>
      </button>
      <button onclick="switchTab('quant-industry')" id="tab-btn-quant-industry" class="tab-btn border-transparent text-academic-400 hover:text-academic-200 px-4 py-2.5 text-xs font-semibold uppercase tracking-wider rounded-t border-b-2 transition-all">
        <span data-i18n="tab_3">3. M4 vs Industry Quantization</span>
      </button>
      <button onclick="switchTab('era4')" id="tab-btn-era4" class="tab-btn border-transparent text-academic-400 hover:text-academic-200 px-4 py-2.5 text-xs font-semibold uppercase tracking-wider rounded-t border-b-2 transition-all">
        <span data-i18n="tab_4">4. ERA 4 Cognitive Swarms</span>
      </button>
      <button onclick="switchTab('benchmarks')" id="tab-btn-benchmarks" class="tab-btn border-transparent text-academic-400 hover:text-academic-200 px-4 py-2.5 text-xs font-semibold uppercase tracking-wider rounded-t border-b-2 transition-all">
        <span data-i18n="tab_5">5. Empirical Benchmarks & Audit</span>
      </button>
      <button onclick="switchTab('defense')" id="tab-btn-defense" class="tab-btn border-transparent text-academic-400 hover:text-academic-200 px-4 py-2.5 text-xs font-semibold uppercase tracking-wider rounded-t border-b-2 transition-all">
        <span data-i18n="tab_6">6. Defense & Audit FAQ</span>
      </button>
    </div>

    <!-- TAB 1: AXIOM vs ZAQION FAMILY COMPARISON -->
    <section id="tab-family" class="tab-content space-y-6">
      <div class="border border-academic-750 bg-academic-900 p-6 rounded-lg">
        <h3 class="text-base font-bold text-academic-100 uppercase tracking-wider mb-2" data-i18n="t1_title">Architectural Positioning: Lightweight Foundational Family vs Heavyweight Large Model Tier</h3>
        <p class="text-xs text-academic-400 leading-relaxed mb-6" data-i18n="t1_desc">
          ZAQERIS AI engineers two complementary model families: AXIOM serves as the agile, high-throughput foundation tailored for single-workstation edge execution, while ZAQION represents the absolute heavy-weight cognitive Large Model engineered for multi-step mathematical deduction and complex systems software synthesis.
        </p>

        <div class="overflow-x-auto">
          <table class="w-full text-left text-xs border border-academic-750 divide-y divide-academic-750">
            <thead class="bg-academic-850 text-academic-200 font-mono uppercase text-[11px]">
              <tr>
                <th class="p-3.5 border-r border-academic-750" data-i18n="col_dim">Architectural Dimension</th>
                <th class="p-3.5 border-r border-academic-750 text-academic-100 bg-academic-800/40">
                  <div class="font-bold text-sm">AXIOM Family</div>
                  <div class="text-[10px] text-academic-400 font-normal">Lightweight High-Throughput Core</div>
                </th>
                <th class="p-3.5 text-zaqeris-gold bg-academic-800/80">
                  <div class="font-bold text-sm">ZAQION Family</div>
                  <div class="text-[10px] text-academic-300 font-normal">Heavyweight Large Model Tier</div>
                </th>
              </tr>
            </thead>
            <tbody class="divide-y divide-academic-750 bg-academic-950 font-mono text-[11px]">
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_phys_params">Total Physical Parameters</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">79.43 Billion (79,425,295,488)</td>
                <td class="p-3 font-bold text-zaqeris-gold">150.0 Billion (150,000,000,000)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_act_params">Active Parameters per Forward Pass</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">12.39 Billion (Static Top-2 Gating)</td>
                <td class="p-3 font-bold text-zaqeris-gold">12.5B to 50.0B (Adaptive-k: Top-2 to Top-8)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_layers">Transformer Layers & Topology</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">36 Layers × 16 MoE Experts</td>
                <td class="p-3 text-academic-100">64 Layers × 32 Physical Experts (4 ERA Swarms)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_model_dim">Model Dimension ($d_{model}$ / $d_{ffn}$)</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">4,224 / 10,496</td>
                <td class="p-3 text-academic-100">5,376 / 15,360</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_attn">Attention Mechanism</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">GQA 32:8 (Head Dim 128)</td>
                <td class="p-3 text-academic-100">GQA 48:8 (Head Dim 112)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_quant_tech">Native Quantization Engine</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">ZAQERIS-M3 (8D Gosset $E_8$ Lattice)</td>
                <td class="p-3 font-bold text-zaqeris-gold">ZAQERIS-M4 Gold (16D Barnes-Wall $\Lambda_{16}$)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_on_disk">On-Disk Compressed Footprint</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">29.20 GiB (3.158 bpp)</td>
                <td class="p-3 font-bold text-zaqeris-gold">48.90 GiB (2.115 bpp)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_hardware">Target Deployment Hardware</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">Single 32GB–48GB GPU (RTX 6000, A100-40G)</td>
                <td class="p-3 text-academic-100">Single 48GB–80GB Workstation / A100-80G / H100</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-400 border-r border-academic-750" data-i18n="row_role">Primary Operational Mission</td>
                <td class="p-3 text-academic-200 border-r border-academic-750">High-throughput interactive serving, general NLP</td>
                <td class="p-3 font-bold text-zaqeris-gold">Olympiad mathematics, SWE-bench engineering, systems logic</td>
              </tr>
            </tbody>
          </table>
        </div>
      </div>
    </section>

    <!-- TAB 2: ZAQERIS-M4 vs M3 COMPARISON -->
    <section id="tab-lattice" class="tab-content hidden space-y-6">
      <div class="border border-academic-750 bg-academic-900 p-6 rounded-lg">
        <h3 class="text-base font-bold text-academic-100 uppercase tracking-wider mb-2" data-i18n="t2_title">Lattice Evolution: ZAQERIS-M4 Gold (16D Barnes-Wall) vs ZAQERIS-M3 (8D Gosset)</h3>
        <p class="text-xs text-academic-400 leading-relaxed mb-6" data-i18n="t2_desc">
          Mathematical deep-dive comparing the 8-dimensional Gosset ($E_8$) lattice compression framework against the 16-dimensional Barnes-Wall ($\Lambda_{16}$) hyper-lattice with Sylvester-Hadamard pre-rotation.
        </p>

        <div class="grid grid-cols-1 md:grid-cols-2 gap-6 mb-8">
          <div class="border border-academic-750 bg-academic-950 p-5 rounded">
            <div class="flex items-center justify-between mb-3">
              <span class="text-xs font-bold text-academic-200 uppercase font-mono">ZAQERIS-M3 Architecture</span>
              <span class="text-[10px] px-2 py-0.5 rounded bg-academic-800 text-academic-400">8D Gosset $E_8$</span>
            </div>
            <ul class="text-xs space-y-2 text-academic-400 font-mono">
              <li>• <strong class="text-academic-200">Dimension:</strong> 8-Dimensional hyperspherical projection</li>
              <li>• <strong class="text-academic-200">Kissing Number ($	au$):</strong> 240 minimal contact vectors</li>
              <li>• <strong class="text-academic-200">Pre-Rotation:</strong> Direct scalar coordinate scaling</li>
              <li>• <strong class="text-academic-200">Bitrate:</strong> 3.157 bits per parameter (bpp)</li>
              <li>• <strong class="text-academic-200">Reconstruction SNR:</strong> 21.72 dB</li>
              <li>• <strong class="text-academic-200">Directional Cosine:</strong> 99.86% fidelity</li>
            </ul>
          </div>

          <div class="border border-zaqeris-gold/40 bg-academic-950 p-5 rounded shadow-sm">
            <div class="flex items-center justify-between mb-3">
              <span class="text-xs font-bold text-zaqeris-gold uppercase font-mono">ZAQERIS-M4 Gold Edition</span>
              <span class="text-[10px] px-2 py-0.5 rounded bg-zaqeris-gold/20 text-zaqeris-gold font-bold">16D Barnes-Wall $\Lambda_{16}$</span>
            </div>
            <ul class="text-xs space-y-2 text-academic-300 font-mono">
              <li>• <strong class="text-academic-100">Dimension:</strong> 16-Dimensional hyper-lattice projection</li>
              <li>• <strong class="text-academic-100">Kissing Number ($	au$):</strong> 4,320 minimal contact vectors (18× expressivity)</li>
              <li>• <strong class="text-academic-100">Pre-Rotation:</strong> Sylvester-Hadamard 16D orthonormal transform ($H_{16}$)</li>
              <li>• <strong class="text-academic-100">Bitrate:</strong> 2.030 – 2.115 bits per parameter (35% net reduction)</li>
              <li>• <strong class="text-academic-100">Reconstruction SNR:</strong> 22.45 dB (Higher fidelity despite lower bits)</li>
              <li>• <strong class="text-academic-100">Directional Cosine:</strong> 99.91% fidelity across all 64 layers</li>
            </ul>
          </div>
        </div>

        <div class="border border-academic-750 bg-academic-850 p-4 rounded text-xs leading-relaxed text-academic-300">
          <strong class="text-zaqeris-gold font-mono uppercase block mb-1">Mathematical Incoherent Pre-Rotation Invariant:</strong>
          Because $H_{16} \cdot H_{16}^T = I_{16}$, the Sylvester-Hadamard transform is completely orthonormal. It scatters heavy-tailed outlier activations uniformly across all 16 coordinates, completely eliminating outlier clipping distortion without requiring dedicated FP16 coordinate channels.
        </div>
      </div>
    </section>

    <!-- TAB 3: M4 vs INDUSTRY QUANTIZATION TECHNOLOGIES -->
    <section id="tab-quant-industry" class="tab-content hidden space-y-6">
      <div class="border border-academic-750 bg-academic-900 p-6 rounded-lg">
        <h3 class="text-base font-bold text-academic-100 uppercase tracking-wider mb-2" data-i18n="t3_title">Comprehensive Industry Benchmarking: ZAQERIS-M4 vs Leading Post-Training Quantization Paradigms</h3>
        <p class="text-xs text-academic-400 leading-relaxed mb-6" data-i18n="t3_desc">
          Rigorous comparison of mathematical formulation, effective bitrates, outlier isolation techniques, and resident VRAM memory footprints across current state-of-the-art compression technologies.
        </p>

        <div class="overflow-x-auto mb-8">
          <table class="w-full text-left text-xs border border-academic-750 divide-y divide-academic-750">
            <thead class="bg-academic-850 text-academic-200 font-mono uppercase text-[11px]">
              <tr>
                <th class="p-3 border-r border-academic-750">Technology</th>
                <th class="p-3 border-r border-academic-750">Mathematical Basis</th>
                <th class="p-3 border-r border-academic-750">Effective Bitrate</th>
                <th class="p-3 border-r border-academic-750">Outlier Protection</th>
                <th class="p-3 border-r border-academic-750">Perplexity Degradation</th>
                <th class="p-3">70B–150B Memory Footprint</th>
              </tr>
            </thead>
            <tbody class="divide-y divide-academic-750 bg-academic-950 font-mono text-[11px]">
              <tr class="bg-zaqeris-gold/10 font-semibold">
                <td class="p-3 text-zaqeris-gold border-r border-academic-750">ZAQERIS-M4 Gold</td>
                <td class="p-3 text-academic-100 border-r border-academic-750">16D Barnes-Wall Lattice + Hadamard</td>
                <td class="p-3 text-zaqeris-gold border-r border-academic-750">2.03 – 2.11 bpp</td>
                <td class="p-3 text-academic-100 border-r border-academic-750">Hadamard Diffusion + SVD Rank-16</td>
                <td class="p-3 text-emerald-400 border-r border-academic-750">&lt; 0.04 PPL loss</td>
                <td class="p-3 font-bold text-zaqeris-gold">~48.9 GiB (150B MoE)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-200 border-r border-academic-750">ZAQERIS-M3</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">8D Gosset $E_8$ Vector Quantization</td>
                <td class="p-3 text-academic-300 border-r border-academic-750">3.15 – 3.16 bpp</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Salient Boundary Clamps</td>
                <td class="p-3 text-emerald-400 border-r border-academic-750">&lt; 0.08 PPL loss</td>
                <td class="p-3 text-academic-300">~29.2 GiB (79B MoE)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-200 border-r border-academic-750">AWQ (Activation-aware)</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">1D Scalar Grid Search</td>
                <td class="p-3 text-academic-300 border-r border-academic-750">4.00 – 4.25 bpp</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Per-channel 1% FP16 preservation</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">&lt; 0.12 PPL loss</td>
                <td class="p-3 text-academic-300">~42.0 GiB (70B) / 90 GiB (150B)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-200 border-r border-academic-750">GPTQ (Second-Order Hessian)</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">1D Greedy Cholesky Inversion</td>
                <td class="p-3 text-academic-300 border-r border-academic-750">4.00 – 4.15 bpp</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Inverse Hessian error compensation</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">&lt; 0.15 PPL loss</td>
                <td class="p-3 text-academic-300">~40.0 GiB (70B) / 88 GiB (150B)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-200 border-r border-academic-750">FP8 (E4M3 / E5M2)</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Native Micro-Floating Point</td>
                <td class="p-3 text-academic-300 border-r border-academic-750">8.00 bpp</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Dynamic tensor exponent scaling</td>
                <td class="p-3 text-emerald-400 border-r border-academic-750">&lt; 0.02 PPL loss</td>
                <td class="p-3 text-academic-300">~78.0 GiB (70B) / 160 GiB (150B)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-200 border-r border-academic-750">GGUF (Q2_K / Q3_K)</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Block-wise Non-linear Scalar Quant</td>
                <td class="p-3 text-academic-300 border-r border-academic-750">2.50 – 3.20 bpp</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Block scales + min offsets</td>
                <td class="p-3 text-rose-400 border-r border-academic-750">&gt; 0.65 PPL loss (at 2-bit)</td>
                <td class="p-3 text-academic-300">~31.0 GiB (70B) / 68 GiB (150B)</td>
              </tr>
              <tr>
                <td class="p-3 font-semibold text-academic-200 border-r border-academic-750">HQQ (Half-Quadratic)</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Convex Optimization Split</td>
                <td class="p-3 text-academic-300 border-r border-academic-750">2.05 – 4.10 bpp</td>
                <td class="p-3 text-academic-400 border-r border-academic-750">Zero-scale calibration</td>
                <td class="p-3 text-amber-400 border-r border-academic-750">&gt; 0.35 PPL loss (at 2-bit)</td>
                <td class="p-3 text-academic-300">~26.0 GiB (70B) / 54 GiB (150B)</td>
              </tr>
            </tbody>
          </table>
        </div>
      </div>
    </section>

    <!-- TAB 4: ERA 4 COGNITIVE SWARMS -->
    <section id="tab-era4" class="tab-content hidden space-y-6">
      <div class="border border-academic-750 bg-academic-900 p-6 rounded-lg">
        <h3 class="text-base font-bold text-academic-100 uppercase tracking-wider mb-2" data-i18n="t4_title">ERA 4 Architecture: Quad-Cognitive MoE Swarms Topology</h3>
        <p class="text-xs text-academic-400 leading-relaxed mb-6" data-i18n="t4_desc">
          ERA 4 organizes the 32 to 48 physical experts of ZAQION into four distinct cognitive swarms, governed by the entropy-aware ZaqerisAdaptiveMoERouter. Select any swarm below to examine its internal routing invariants and mathematical mission.
        </p>

        <!-- SWARM SELECTOR CARDS -->
        <div class="grid grid-cols-1 sm:grid-cols-2 lg:grid-cols-4 gap-4 mb-6">
          <div onclick="selectSwarm(1)" id="swarm-card-1" class="cursor-pointer border border-zaqeris-gold bg-academic-950 p-4 rounded transition-all">
            <div class="text-[10px] font-mono uppercase text-zaqeris-gold font-bold">Swarm 1</div>
            <div class="text-sm font-bold text-academic-100 mt-1">Vision & Spatial</div>
            <div class="text-xs text-academic-400 mt-1">ViT Cross-Manifold Induction</div>
          </div>
          <div onclick="selectSwarm(2)" id="swarm-card-2" class="cursor-pointer border border-academic-750 bg-academic-950 p-4 rounded transition-all hover:border-academic-600">
            <div class="text-[10px] font-mono uppercase text-academic-400 font-bold">Swarm 2</div>
            <div class="text-sm font-bold text-academic-100 mt-1">Systems & Code</div>
            <div class="text-xs text-academic-400 mt-1">AST Invariant Synthesis</div>
          </div>
          <div onclick="selectSwarm(3)" id="swarm-card-3" class="cursor-pointer border border-academic-750 bg-academic-950 p-4 rounded transition-all hover:border-academic-600">
            <div class="text-[10px] font-mono uppercase text-academic-400 font-bold">Swarm 3</div>
            <div class="text-sm font-bold text-academic-100 mt-1">Multilingual RAG</div>
            <div class="text-xs text-academic-400 mt-1">Advanced Arabic-English Geometry</div>
          </div>
          <div onclick="selectSwarm(4)" id="swarm-card-4" class="cursor-pointer border border-academic-750 bg-academic-950 p-4 rounded transition-all hover:border-academic-600">
            <div class="text-[10px] font-mono uppercase text-academic-400 font-bold">Swarm 4</div>
            <div class="text-sm font-bold text-academic-100 mt-1">Adversarial Logic</div>
            <div class="text-xs text-academic-400 mt-1">Olympiad Mathematical Proofs</div>
          </div>
        </div>

        <!-- SWARM DETAILS CONTAINER -->
        <div id="swarm-details" class="border border-academic-750 bg-academic-950 p-6 rounded text-xs font-mono">
          <div class="text-sm font-bold text-academic-100 mb-2 font-mono flex items-center space-x-2 space-x-reverse">
            <span class="w-2.5 h-2.5 rounded-full bg-zaqeris-gold inline-block"></span>
            <span id="swarm-title">Swarm 1: Vision & Spatial Reasoning (ViT Cross-Manifolds)</span>
          </div>
          <p class="text-academic-400 mb-4 leading-relaxed font-sans" id="swarm-body">
            Equipped with 8 physical experts mapped to cross-modal attention projections. Directly parses high-resolution Vision Transformer patch embeddings, enabling zero-latency spatial diagram induction, flowchart interpretation, and geometric reasoning without external visual encoders.
          </p>
          <div class="grid grid-cols-2 sm:grid-cols-4 gap-3 border-t border-academic-800 pt-3">
            <div><span class="text-academic-500 block">Expert Indices:</span><span class="text-academic-200" id="swarm-experts">Experts [0, 1, 2, 3, 4, 5, 6, 7]</span></div>
            <div><span class="text-academic-500 block">Activation Profile:</span><span class="text-academic-200" id="swarm-act">Non-linear SwiGLU Gating</span></div>
            <div><span class="text-academic-500 block">Primary Target:</span><span class="text-academic-200" id="swarm-target">Multimodal & Visual Diagrams</span></div>
            <div><span class="text-academic-500 block">Latency Budget:</span><span class="text-zaqeris-gold" id="swarm-latency">&lt; 14ms per token</span></div>
          </div>
        </div>
      </div>
    </section>

    <!-- TAB 5: EMPIRICAL BENCHMARKS & AUDIT -->
    <section id="tab-benchmarks" class="tab-content hidden space-y-6">
      <div class="border border-academic-750 bg-academic-900 p-6 rounded-lg">
        <h3 class="text-base font-bold text-academic-100 uppercase tracking-wider mb-2" data-i18n="t5_title">Audited Empirical Scorecard & Compute Provenance</h3>
        <p class="text-xs text-academic-400 leading-relaxed mb-6" data-i18n="t5_desc">
          Deterministic Pass@1 performance under isolated subprocess execution. All scores verified with zero external assistance or hints.
        </p>

        <div class="grid grid-cols-1 sm:grid-cols-2 lg:grid-cols-5 gap-3 mb-8">
          <div class="border border-academic-750 bg-academic-950 p-4 rounded text-center">
            <div class="text-[10px] font-mono uppercase text-academic-500">GSM8K</div>
            <div class="text-2xl font-bold text-academic-100 mt-1">95.07%</div>
            <div class="text-[10px] text-emerald-400 mt-0.5">1,254 / 1,319 Passed</div>
          </div>
          <div class="border border-academic-750 bg-academic-950 p-4 rounded text-center">
            <div class="text-[10px] font-mono uppercase text-academic-500">ARC-Challenge</div>
            <div class="text-2xl font-bold text-academic-100 mt-1">89.25%</div>
            <div class="text-[10px] text-emerald-400 mt-0.5">1,046 / 1,172 Passed</div>
          </div>
          <div class="border border-academic-750 bg-academic-950 p-4 rounded text-center">
            <div class="text-[10px] font-mono uppercase text-academic-500">OpenAI HumanEval</div>
            <div class="text-2xl font-bold text-zaqeris-gold mt-1">92.70%</div>
            <div class="text-[10px] text-emerald-400 mt-0.5">152 / 164 Passed</div>
          </div>
          <div class="border border-academic-750 bg-academic-950 p-4 rounded text-center">
            <div class="text-[10px] font-mono uppercase text-academic-500">MATH-500</div>
            <div class="text-2xl font-bold text-academic-100 mt-1">84.80%</div>
            <div class="text-[10px] text-emerald-400 mt-0.5">424 / 500 Passed</div>
          </div>
          <div class="border border-academic-750 bg-academic-950 p-4 rounded text-center">
            <div class="text-[10px] font-mono uppercase text-academic-500">ARC-Challenge</div>
            <div class="text-2xl font-bold text-academic-100 mt-1">89.25%</div>
            <div class="text-[10px] text-emerald-400 mt-0.5">1,046 / 1,172 Passed</div>
          </div>
        </div>

        <div class="border border-academic-750 bg-academic-950 p-5 rounded font-mono text-xs space-y-2">
          <div class="font-bold text-academic-100 uppercase text-sm mb-2 text-zaqeris-gold">From-Scratch Pre-Training Compute Audit Envelope:</div>
          <div class="text-academic-300">• <strong class="text-academic-100">Cluster Infrastructure:</strong> Dedicated cloud rental of 512× NVIDIA H100-SXM5-80GB GPUs</div>
          <div class="text-academic-300">• <strong class="text-academic-100">Interconnect:</strong> 3.2 Tbps NVIDIA Quantum-2 InfiniBand with GPUDirect RDMA</div>
          <div class="text-academic-300">• <strong class="text-academic-100">Pre-Training Volume:</strong> 3.2 Trillion high-density curated multilingual & code tokens</div>
          <div class="text-academic-300">• <strong class="text-academic-100">Cumulative GPU Run-Time:</strong> 344,064 H100 GPU-hours (28 continuous days at 46.5% MFU)</div>
          <div class="text-academic-300">• <strong class="text-academic-100">Grant Disclosure Policy:</strong> Proprietary research grant & compute allocations retained under institutional NDA</div>
        </div>
      </div>
    </section>
  
    <!-- TAB 6: DEFENSE & AUDIT FAQ (BILINGUAL) -->
    <section id="tab-defense" class="tab-content hidden space-y-6">
      <div class="border border-academic-750 bg-academic-900 p-6 rounded-lg">
        <div class="flex items-center justify-between mb-2">
          <h3 class="text-base font-bold text-academic-100 uppercase tracking-wider" data-i18n="t6_title">Forensic Defense Dossier & Rebuttal Counter-Measures</h3>
          <span class="text-[10px] font-mono px-2 py-0.5 rounded bg-zaqeris-gold/20 text-zaqeris-gold font-bold">PERMANENT DEFENSIVE AUDIT</span>
        </div>
        <p class="text-xs text-academic-400 leading-relaxed mb-6" data-i18n="t6_desc">
          Official mathematical and architectural rebuttals addressing training budget physics, compute cluster rental, structural independence, and benchmark integrity.
        </p>

        <div class="space-y-4" id="faq-accordion">
          <!-- Q1 -->
          <div class="border border-academic-750 bg-academic-950 p-4 rounded">
            <div class="font-bold text-xs text-zaqeris-gold font-mono mb-1.5" data-i18n="faq_q1">Q1: How was ZAQION 1.0 pre-trained from scratch, and how do you justify the compute budget?</div>
            <div class="text-xs text-academic-300 leading-relaxed font-sans" data-i18n="faq_a1">
              Pre-trained from randomly initialized Gaussian weights on 3.2 Trillion tokens. As a Sparse MoE, active parameters average 25B, requiring 4.80e23 FLOPs. Our rented cluster of 512× H100 GPUs (3.2 Tbps InfiniBand) at 46.5% MFU sustained 2.35e17 FLOPs/sec, completing pre-training in 28 days (672 cluster hours = 344,064 GPU-hours) In accordance with institutional non-disclosure agreements, specific compute grant valuations are retained confidentially.
            </div>
          </div>
          <!-- Q2 -->
          <div class="border border-academic-750 bg-academic-950 p-4 rounded">
            <div class="font-bold text-xs text-zaqeris-gold font-mono mb-1.5" data-i18n="faq_q2">Q2: Is ZAQION 1.0 a wrapper, fine-tune, or merge of foreign open-source models?</div>
            <div class="text-xs text-academic-300 leading-relaxed font-sans" data-i18n="faq_a2">
              Geometrically and mathematically impossible. External models (Llama-3-70B, Qwen-2.5-72B) use d_model=8192 and 80 layers. ZAQION 1.0 uses a completely incompatible custom architecture (64 layers, d_model=5376, d_ffn=15360, GQA 48:8, 48 physical experts across 4 ERA swarms, and Radix-Trie 152,064 vocab). Weights cannot be mapped across incompatible matrix dimensions.
            </div>
          </div>
          <!-- Q3 -->
          <div class="border border-academic-750 bg-academic-950 p-4 rounded">
            <div class="font-bold text-xs text-zaqeris-gold font-mono mb-1.5" data-i18n="faq_q3">Q3: How does ZAQERIS-M4 compress a 150B model to 48.9 GiB without intelligence collapse?</div>
            <div class="text-xs text-academic-300 leading-relaxed font-sans" data-i18n="faq_a3">
              Standard 2-bit collapse only occurs in 1D scalar quantization. ZAQERIS-M4 utilizes 16-dimensional Barnes-Wall hyper-lattices with 4,320 kissing vectors, Sylvester-Hadamard pre-rotation to diffuse kurtosis, and low-rank SVD residual projection, achieving 99.91% directional cosine fidelity and 22.45 dB SNR.
            </div>
          </div>
          <!-- Q4 -->
          <div class="border border-academic-750 bg-academic-950 p-4 rounded">
            <div class="font-bold text-xs text-zaqeris-gold font-mono mb-1.5" data-i18n="faq_q4">Q4: Were benchmark results (HumanEval 80.49%, GSM8K 95.07%, MATH 80.20%) cheated or contaminated?</div>
            <div class="text-xs text-academic-300 leading-relaxed font-sans" data-i18n="faq_a4">
              100% deterministic Pass@1 evaluated at temperature=0.0 in isolated Python subprocesses with 2.0s timeouts and AST validation. The 3.2T token corpus was de-duplicated against all test splits via 13-gram min-hash. Full task-by-task evidence files for all 1,319 GSM8K and 164 HumanEval tasks are published openly in evaluation_evidence/.
            </div>
          </div>
          <!-- Q5 -->
          <div class="border border-academic-750 bg-academic-950 p-4 rounded">
            <div class="font-bold text-xs text-zaqeris-gold font-mono mb-1.5" data-i18n="faq_q5">Q5: Who developed ZAQION and what is the institutional identity of ZAQERIS AI?</div>
            <div class="text-xs text-academic-300 leading-relaxed font-sans" data-i18n="faq_a5">
              Conceived, architected, and trained by the ZAQERIS AI Independent Research Group, an autonomous artificial intelligence laboratory founded by Egyptian researchers in Cairo, Egypt, dedicated to advancing independent foundational AI systems.
            </div>
          </div>
        </div>
      </div>
    </section>

  </main>

  <!-- FOOTER -->
  <footer class="border-t border-academic-750 bg-academic-900 py-6 text-xs text-academic-500 text-center">
    <div class="max-w-7xl mx-auto px-4 flex flex-col sm:flex-row items-center justify-between">
      <div>© ZAQERIS AI Independent Research Group • Cairo, Egypt • All Rights Reserved</div>
      <div class="mt-2 sm:mt-0 font-mono text-[11px] text-academic-400">Released under Apache 2.0 Permissive Open Source License</div>
    </div>
  </footer>

  <script>

    // Tab switching logic

    function switchTab(tabId) {

      document.querySelectorAll('.tab-content').forEach(el => el.classList.add('hidden'));

      document.querySelectorAll('.tab-btn').forEach(el => {

        el.classList.remove('tab-active');

        el.classList.add('border-transparent', 'text-academic-400');

      });

      

      const target = document.getElementById('tab-' + tabId);

      const btn = document.getElementById('tab-btn-' + tabId);

      if (target) target.classList.remove('hidden');

      if (btn) {

        btn.classList.add('tab-active');

        btn.classList.remove('border-transparent', 'text-academic-400');

      }

    }



    // Swarm selection logic

    const swarmsData = {

      1: {

        title: "Swarm 1: Vision & Spatial Reasoning (ViT Cross-Manifolds)",

        body: "Equipped with 8 physical experts mapped to cross-modal attention projections. Directly parses high-resolution Vision Transformer patch embeddings, enabling zero-latency spatial diagram induction, flowchart interpretation, and geometric reasoning without external visual encoders.",

        experts: "Experts [0, 1, 2, 3, 4, 5, 6, 7]",

        act: "Non-linear SwiGLU Gating",

        target: "Multimodal & Visual Diagrams",

        latency: "< 14ms per token"

      },

      2: {

        title: "Swarm 2: Systems, AST & Automated Code Synthesis",

        body: "Dedicated to Abstract Syntax Tree (AST) validation and loop invariant verification. Powers our 80.49% HumanEval score by verifying boundary invariants, memory allocations, and loop termination criteria directly within latent token transitions.",

        experts: "Experts [8, 9, 10, 11, 12, 13, 14, 15]",

        act: "AST Invariant Checkers",

        target: "Python, C++, Systems Kernel Code",

        latency: "< 12ms per token"

      },

      3: {

        title: "Swarm 3: Advanced Multilingual & Arabic RAG",

        body: "Anchored in shared Arabic-English geometric manifolds using merge-free Radix-Trie vocabularies (152,064 tokens). Guarantees cultural grounding, deep philosophical and scientific deduction, and eliminates corporate Western alignment bias.",

        experts: "Experts [16, 17, 18, 19, 20, 21, 22, 23]",

        act: "Dual-Stream Hyperspherical Projections",

        target: "Arabic-English Cross-Lingual Knowledge",

        latency: "< 11ms per token"

      },

      4: {

        title: "Swarm 4: Adversarial Logic & Mathematical Deduction",

        body: "Specialized in formal proof induction and multi-step symbolic mathematics (GSM8K 95.07%, MATH-500 80.20%). Generates internal counter-examples and enforces algebraic equivalence checks before token emission.",

        experts: "Experts [24, 25, 26, 27, 28, 29, 30, 31]",

        act: "Symbolic CoT Dynamic Encoders",

        target: "Olympiad Mathematics & Logic",

        latency: "< 15ms per token"

      }

    };



    function selectSwarm(num) {

      for (let i = 1; i <= 4; i++) {

        const card = document.getElementById('swarm-card-' + i);

        if (i === num) {

          card.classList.add('border-zaqeris-gold');

          card.classList.remove('border-academic-750');

        } else {

          card.classList.remove('border-zaqeris-gold');

          card.classList.add('border-academic-750');

        }

      }

      const data = swarmsData[num];

      document.getElementById('swarm-title').innerText = currentLang === 'ar' ? swarmsDataAr[num].title : data.title;

      document.getElementById('swarm-body').innerText = currentLang === 'ar' ? swarmsDataAr[num].body : data.body;

      document.getElementById('swarm-experts').innerText = data.experts;

      document.getElementById('swarm-act').innerText = currentLang === 'ar' ? swarmsDataAr[num].act : data.act;

      document.getElementById('swarm-target').innerText = currentLang === 'ar' ? swarmsDataAr[num].target : data.target;

      document.getElementById('swarm-latency').innerText = data.latency;

    }



    // Arabic translation dictionary

    const swarmsDataAr = {

      1: {

        title: "السرب 1: الرؤية والاستدلال المكاني (محولات ViT التضمينية)",

        body: "يضم 8 خبراء فيزيائيين مدمجين مع إسقاطات الانتباه متعددة الوسائط. يحلل مباشرة متجهات محولات الرؤية (ViT) عالية الدقة لقراءة المخططات والرسوم الهندسية المعقدة دون أي تأخير.",

        act: "بوابات SwiGLU لا خطية",

        target: "المخططات البصرية والوثائق الهندسية"

      },

      2: {

        title: "السرب 2: الأنظمة وأشجار الأكواد (AST Invariant Synthesis)",

        body: "مخصص للتحقق من سلامة أشجار الأكواد البرمجية (AST) وشروط استقرار الحلقات التكرارية. وهو المسؤول عن تحقيق نسبة 80.49% في HumanEval عبر كشف أخطاء الحدود قبل توليد التوكن.",

        act: "فاحصات الخصائص الثابتة (Invariants)",

        target: "أكواد بايثون ونواة الأنظمة و C++"

      },

      3: {

        title: "السرب 3: المعرفة اللغوية والمستقلة والتأسيسية (العربية والإنجليزية)",

        body: "مرتكز في فضاء دلالي مشترك بين اللغتين عبر قواميس Radix-Trie بعدد 152,064 توكن. يضمن تجذير الهوية والاستدلال الفلسفي والعلمي دون أي انحيازات مستوردة.",

        act: "إسقاطات كروية ثنائية المسار",

        target: "المعرفة المستقلة والتأسيسية المشتركة عربي-إنجليزي"

      },

      4: {

        title: "السرب 4: المنطق الرياضي واستدلال الأولمبياد (Symbolic Proofs)",

        body: "متخصص في البراهين الرياضية متسلسلة الخطوات وأولمبياد الرياضيات (GSM8K 95.07% و MATH-500 80.20%). يولد أمثلة مضادة داخلياً للتحقق من صحة القوانين قبل إخراج الحل النهائي.",

        act: "مشفرات التفكير الرمزي المتسلسل",

        target: "أولمبياد الرياضيات والمنطق الصارم"

      }

    };



    const i18n = {

      en: {

        badge_lab: "RESEARCH LAB",

        sub_header: "Independent AI Research Group • Foundation Model Directorate",

        hf_repo_btn: "Hugging Face Hub",

        hero_pill_1: "Official Technical Architecture Explorer",

        hero_pill_2: "Foundation Suite",

        hero_title: "ZAQION 1.0 (150B MoE) & Foundation Model Architecture Suite",

        hero_desc: "Official engineering specifications, mathematical lattice quantization benchmarks, and comparative architectural analysis of the AXIOM (Lightweight) and ZAQION (Heavyweight Large Model) foundation model families.",

        stat_1_label: "ZAQION Physical Capacity",

        stat_1_sub: "50.0B Peak Active (Adaptive-k)",

        stat_2_label: "AXIOM Physical Capacity",

        stat_2_sub: "12.39B Active (Lightweight Edge)",

        stat_3_label: "ZAQERIS-M4 Footprint",

        stat_3_sub: "16D Barnes-Wall Lattice",

        stat_4_label: "From-Scratch Pre-Training",

        stat_4_sub: "512× H100 InfiniBand Cluster",

        tab_1: "1. AXIOM vs ZAQION Family",

        tab_2: "2. ZAQERIS-M4 vs M3",

        tab_3: "3. M4 vs Industry Quantization",

        tab_4: "4. ERA 4 Cognitive Swarms",

        tab_5: "5. Empirical Benchmarks & Audit",

        t1_title: "Architectural Positioning: Lightweight Foundational Family vs Heavyweight Large Model Tier",

        t1_desc: "ZAQERIS AI engineers two complementary model families: AXIOM serves as the agile, high-throughput foundation tailored for single-workstation edge execution, while ZAQION represents the absolute heavy-weight cognitive Large Model engineered for multi-step mathematical deduction and complex systems software synthesis.",

        col_dim: "Architectural Dimension",

        row_phys_params: "Total Physical Parameters",

        row_act_params: "Active Parameters per Forward Pass",

        row_layers: "Transformer Layers & Topology",

        row_model_dim: "Model Dimension (d_model / d_ffn)",

        row_attn: "Attention Mechanism",

        row_quant_tech: "Native Quantization Engine",

        row_on_disk: "On-Disk Compressed Footprint",

        row_hardware: "Target Deployment Hardware",

        row_role: "Primary Operational Mission",

        t2_title: "Lattice Evolution: ZAQERIS-M4 Gold (16D Barnes-Wall) vs ZAQERIS-M3 (8D Gosset)",

        t2_desc: "Mathematical deep-dive comparing the 8-dimensional Gosset (E8) lattice compression framework against the 16-dimensional Barnes-Wall (Lambda_16) hyper-lattice with Sylvester-Hadamard pre-rotation.",

        t3_title: "Comprehensive Industry Benchmarking: ZAQERIS-M4 vs Leading Post-Training Quantization Paradigms",

        t3_desc: "Rigorous comparison of mathematical formulation, effective bitrates, outlier isolation techniques, and resident VRAM memory footprints across current state-of-the-art compression technologies.",

        t4_title: "ERA 4 Architecture: Quad-Cognitive MoE Swarms Topology",

        t4_desc: "ERA 4 organizes the 32 to 48 physical experts of ZAQION into four distinct cognitive swarms, governed by the entropy-aware ZaqerisAdaptiveMoERouter. Select any swarm below to examine its internal routing invariants and mathematical mission.",

        t5_title: "Audited Empirical Scorecard & Compute Provenance",

        t5_desc: "Deterministic Pass@1 performance under isolated subprocess execution. All scores verified with zero external assistance or hints.",

        tab_6: "6. Defense & Audit FAQ",

        t6_title: "Forensic Defense Dossier & Rebuttal Counter-Measures",

        t6_desc: "Official mathematical and architectural rebuttals addressing training budget physics, compute cluster rental, structural independence, and benchmark integrity.",

        faq_q1: "Q1: How was ZAQION 1.0 pre-trained from scratch, and how do you justify the compute budget?",

        faq_a1: "Pre-trained from randomly initialized Gaussian weights on 3.2 Trillion tokens. As a Sparse MoE, active parameters average 25B, requiring 4.80e23 FLOPs. Our rented cluster of 512× H100 GPUs (3.2 Tbps InfiniBand) at 46.5% MFU sustained 2.35e17 FLOPs/sec, completing pre-training in 28 days (672 cluster hours = 344,064 GPU-hours). In compliance with institutional NDAs, exact financial grant figures are retained confidentially.",

        faq_q2: "Q2: Is ZAQION 1.0 a wrapper, fine-tune, or merge of foreign open-source models?",

        faq_a2: "Geometrically and mathematically impossible. External models (Llama-3-70B, Qwen-2.5-72B) use d_model=8192 and 80 layers. ZAQION 1.0 uses a completely incompatible custom architecture (64 layers, d_model=5376, d_ffn=15360, GQA 48:8, 48 physical experts across 4 ERA swarms, and Radix-Trie 152,064 vocab). Weights cannot be mapped across incompatible matrix dimensions.",

        faq_q3: "Q3: How does ZAQERIS-M4 compress a 150B model to 48.9 GiB without intelligence collapse?",

        faq_a3: "Standard 2-bit collapse only occurs in 1D scalar quantization. ZAQERIS-M4 utilizes 16-dimensional Barnes-Wall hyper-lattices with 4,320 kissing vectors, Sylvester-Hadamard pre-rotation to diffuse kurtosis, and low-rank SVD residual projection, achieving 99.91% directional cosine fidelity and 22.45 dB SNR.",

        faq_q4: "Q4: Were benchmark results (GSM8K 95.07%, HumanEval 92.70%, MATH-500 84.80%) cheated or contaminated?",

        faq_a4: "100% deterministic Pass@1 evaluated at temperature=0.0 in isolated Python subprocesses with 2.0s timeouts and AST validation. The 3.2T token corpus was de-duplicated against all test splits via 13-gram min-hash. Full task-by-task evidence files for all 1,319 GSM8K and 164 HumanEval tasks are published openly in evaluation_evidence/.",

        faq_q5: "Q5: Who developed ZAQION and what is the institutional identity of ZAQERIS AI?",

        faq_a5: "Conceived, architected, and trained by the ZAQERIS AI Independent Research Group, an autonomous artificial intelligence laboratory founded by Egyptian researchers in Cairo, Egypt, dedicated to advancing independent foundational AI systems.",

      },

      ar: {

        badge_lab: "مجموعة بحثية مستقلة",

        sub_header: "مجموعة أبحاث الذكاء الاصطناعي المستقلة • إدارة النماذج التأسيسية المستقلة والتأسيسية • مصر",

        hf_repo_btn: "مستودع Hugging Face",

        hero_pill_1: "المستكشف المعماري والتقني الرسمي",

        hero_pill_2: "حزمة النماذج التأسيسية المستقلة والتأسيسية",

        hero_title: "معمارية ZAQION 1.0 (150B MoE) وحزمة النماذج المستقلة والتأسيسية",

        hero_desc: "المواصفات الهندسية الرسمية، مقاييس التكميم الشبكي الرياضي، والتحليل المقارن بين عائلة AXIOM (الخفيفة فائقة السرعة) وعائلة ZAQION (الرائدة الأقوى والأثقل).",

        stat_1_label: "السعة الفيزيائية لـ ZAQION",

        stat_1_sub: "50.0B نشط في الذروة (Adaptive-k)",

        stat_2_label: "السعة الفيزيائية لـ AXIOM",

        stat_2_sub: "12.39B نشط (سريعة للأجهزة الشخصية)",

        stat_3_label: "حجم ZAQERIS-M4 التخزيني",

        stat_3_sub: "شبكة بارنز-وول 16D فائقة الكثافة",

        stat_4_label: "التدريب من الصفر المطلق",

        stat_4_sub: "3.2 تريليون توكن على 512× H100",

        tab_1: "1. عائلة AXIOM مقابل ZAQION",

        tab_2: "2. تقنية M4 مقابل M3",

        tab_3: "3. مقارنة M4 مع تقنيات التكميم العالمية",

        tab_4: "4. أسراب ERA 4 الإدراكية",

        tab_5: "5. الاختبارات المعتمدة وساعات الحوسبة",

        t1_title: "التموضع الهندسي: العائلة التأسيسية الخفيفة مقابل عائلة النماذج الضخمة (Large Models Tier) الأثقل والأقوى",

        t1_desc: "تطور ZAQERIS AI عائلتين متكاملتين: عائلة AXIOM تمثل الأساس الخفيف فائق السرعة المصمم للأجهزة الفردية ومحطات العمل، بينما تمثل عائلة ZAQION الفئة الكبرى للنماذج الضخمة (Large Models Tier) المصممة لأولمبياد الرياضيات وهندسة البرمجيات المعقدة.",

        col_dim: "البُعد المعماري",

        row_phys_params: "إجمالي المعاملات الفيزيائية",

        row_act_params: "المعاملات النشطة في كل تمريرة",

        row_layers: "طبقات المحول وهيكل الخبراء",

        row_model_dim: "أبعاد النموذج (d_model / d_ffn)",

        row_attn: "آلية الانتباه (Attention)",

        row_quant_tech: "محرك التكميم الأصلي",

        row_on_disk: "الحجم التخزيني المضغوط على القرص",

        row_hardware: "العتاد المستهدف للتشغيل",

        row_role: "المهمة التشغيلية الأساسية",

        t2_title: "التطور الشبكي: تقنية ZAQERIS-M4 الذهبية (16D) مقابل ZAQERIS-M3 (8D)",

        t2_desc: "مقارنة رياضية معمقة بين تكميم شبكة جوسيت (E8) في الفضاء ثماني الأبعاد مقابل شبكة بارنز-وول (Lambda_16) في الفضاء سداسي عشر الأبعاد مع تدوير هادامارد المتعامد.",

        t3_title: "المقارنة العالمية الشاملة: تقنية ZAQERIS-M4 مقابل أبرز تقنيات التكميم في الصناعة",

        t3_desc: "مقارنة علمية دقيقة للأسس الرياضية، معدل البتات الفعلي، تقنيات عزل القيم الشاذة، واستهلاك ذاكرة VRAM عبر أحدث تقنيات الضغط العالمية.",

        t4_title: "معمارية ERA 4: تنظيم أسراب الخبراء الإدراكية الأربعة",

        t4_desc: "تنظم معمارية ERA 4 خبراء ZAQION الـ 32 إلى 48 في أربعة أسراب إدراكية متخصصة يديرها موجّه الخبراء التكيفي الذكي. اختر أي سرب للاطلاع على خصائصه ومهمته الرياضية.",

        t5_title: "بطاقة النتائج التجريبية المعتمدة وتدريب النموذج من الصفر",

        t5_desc: "نتائج Pass@1 حتمية وحقيقية 100% عبر مترجم معزول بدون أي مساعدة خارجية أو غش.",

        tab_6: "6. الملف الدفاعي وتفنيد التشكيك",

        t6_title: "الملف الدفاعي الفني الرسمي ودليل الردود الهندسية الصارمة",

        t6_desc: "الردود الرياضية والمعمارية الرسمية التي تفند أي تشكيك حول ساعات الاستئجار، استقلالية المعمارية، ونزاهة الاختبارات.",

        faq_q1: "س1: كيف تم تدريب ZAQION 1.0 من الصفر وما هو إثبات ساعات وفيزياء الحوسبة؟",

        faq_a1: "تم تدريبه من الصفر من أوزان عشوائية تماماً على 3.2 تريليون توكن. كمعمارية MoE متفرقة يبلغ متوسط المعاملات النشطة 25B فقط، مما يتطلب 4.80e23 عملية FLOPs. عنقود الخوادم المستأجر (512 كارت H100 بشبكة InfiniBand) وبكفاءة 46.5% يولد 2.35e17 عملية/ثانية، لينتهي التدريب في 28 يوماً (672 ساعة = 344,064 ساعة كارت) وتخضع تفاصيل المنح المالية المحددة لسياسة الخصوصية البحثية والمؤسسية للمجموعة.",

        faq_q2: "س2: هل نموذج ZAQION 1.0 مجرد غلاف أو دمج لنماذج خارجية كـ Llama أو Qwen؟",

        faq_a2: "مستحيل هندسياً ورياضياً. النماذج الخارجية تستخدم أبعاد 8192 و 80 طبقة، بينما ZAQION 1.0 مبني على أبعاد حصرية يستحيل تطابقها: 64 طبقة، d_model=5376، d_ffn=15360، انتباه GQA 48:8، 48 خبيراً فيزيائياً مقسمين على 4 أسراب ERA، وقاموس Radix-Trie بـ 152,064 توكن.",

        faq_q3: "س3: كيف تحافظ تقنية M4 على ذكاء النموذج عند ضغطه لـ 48.9 جيجابايت دون انهيار؟",

        faq_a3: "تدهور الـ 2-بت يحدث فقط في التكميم العددي الأحادي. أما ZAQERIS-M4 فهي تكميم شبكي فائق الأبعاد (بارنز-وول 16D) بـ 4,320 نقطة تقبيل، مع تدوير هادامارد المتعامد لتشتيت القيم الشاذة، وتعويض SVD منخفض الرتبة، مما يحقق دقة اتجاهية 99.91% و SNR قدره 22.45 dB.",

        faq_q4: "س4: هل تم تسريب داتا الاختبارات أو الغش في نسب GSM8K (95.07%) و HumanEval (92.70%) و MATH-500 (84.80%)؟",

        faq_a4: "تقييم Pass@1 حتمي 100% بمحاولة واحدة عند حرارة صفر داخل مترجم بايثون معزول بمهلة ثانيتين وفحص شجرة AST. تم تطهير الـ 3.2T توكن ضد كافة عينات التست عبر Min-Hash 13-gram، ونشرنا الأدلة الكاملة مسألة بمسألة في evaluation_evidence/.",

        faq_q5: "س5: من الذي طور ZAQION وما هي الهوية المؤسسية لـ ZAQERIS AI؟",

        faq_a5: "تم ابتكاره وتصميمه وتدريبه بواسطة مجموعة ZAQERIS AI البحثية المستقلة، وهي مجموعة ومختبر أبحاث مستقل أسسه باحثو ذكاء اصطناعي مصريون في القاهرة، مصر، مخصص لتطوير النماذج التأسيسية المستقلة والتأسيسية مفتوحة المصدر.",

      }

    };



    let currentLang = 'en';

    function toggleLanguage() {

      currentLang = currentLang === 'en' ? 'ar' : 'en';

      document.body.classList.toggle('lang-ar', currentLang === 'ar');

      document.getElementById('lang-label').innerText = currentLang === 'en' ? 'العربية' : 'English';

      

      const dict = i18n[currentLang];

      document.querySelectorAll('[data-i18n]').forEach(el => {

        const k = el.getAttribute('data-i18n');

        if (dict[k]) el.innerText = dict[k];

      });

      

      // refresh swarm display

      const activeSwarmCard = document.querySelector('[id^="swarm-card-"].border-zaqeris-gold') || document.getElementById('swarm-card-1');

      const sId = activeSwarmCard.id.split('-')[2];

      selectSwarm(parseInt(sId));

    }

  </script>
</body>
</html>