Hugging Face's logo

3kmfi6hp
/
temp-scripts

Model card Files
xet
 Community
temp-scripts
File size: 3,012 Bytes
3111a76
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
#!/usr/bin/env python3
import urllib.request, json
from datetime import datetime, timezone

url = 'https://llmsnare-api.mistermorph.com/v1/timelines?limit=96'
data = json.loads(urllib.request.urlopen(url, timeout=60).read())

NAMES = {
    'deepseek_v4_pro': 'DeepSeek V4 Pro',
    'deepseek_v4_flash': 'DeepSeek V4 Flash',
    'claude_opus_4_6': 'Claude Opus 4.6',
    'claude_opus_4_7': 'Claude Opus 4.7',
    'claude_sonnet_4_6': 'Claude Sonnet 4.6',
    'openai_gpt_5_5': 'GPT-5.5',
    'openai_gpt_5_4': 'GPT-5.4',
    'minimax_m_2_7': 'MiniMax M2.7',
    'minimax_m_2_5': 'MiniMax M2.5',
    'glm_5_turbo': 'GLM-5 Turbo',
    'glm_5_1': 'GLM-5.1',
    'grok_4_1_fast_reasoning': 'Grok 4.1 Fast',
    'grok_4_2_reasoning': 'Grok 4.2',
    'gemini_3_5_flash': 'Gemini 3.5 Flash',
    'gemini_3_1_pro_preview': 'Gemini 3.1 Pro',
    'gemini_3_1_flash_lite': 'Gemini 3.1 Flash Lite',
    'gemini_2_5_flash': 'Gemini 2.5 Flash',
    'kimi_k_2_5': 'Kimi K2.5',
    'kimi_k_2_6': 'Kimi K2.6',
    'openrouter_qwen_3_5_27b': 'Qwen 3.5 27B',
    'openrouter_qwen_3_6_plus': 'Qwen 3.6 Plus',
    'openrouter_gemma_4_31b': 'Gemma 4 31B',
    'openrouter_mimo_v2_pro': 'Mimo V2 Pro'
}

rows = []
for pid, pdata in data.get('profiles', {}).items():
    entries = pdata.get('entries', [])
    name = NAMES.get(pid, pid)
    points = sorted([(e['timestamp'], e['normalized_score'], e.get('raw_score',0), e.get('max_score',0)) for e in entries], key=lambda x: x[0])
    scores = [p[1] for p in points]
    raw = [p[2] for p in points]
    maxs = [p[3] for p in points]
    n = len(scores)
    if n < 4:
        continue
    avg8 = round(sum(scores[-8:]) / 8, 1)
    prev8 = scores[-16:-8] if n >= 16 else scores[:max(1,n-8)]
    avg_prev8 = round(sum(prev8)/len(prev8),1) if prev8 else avg8
    change = round(avg8 - avg_prev8, 1)
    latest = round(scores[-1], 1)
    latest_raw = int(raw[-1])
    latest_max = int(maxs[-1])
    rows.append({'name':name,'avg8':avg8,'change':change,'latest':latest,'raw':latest_raw,'max':latest_max})

rows.sort(key=lambda x: -x['avg8'])
now = datetime.now(timezone.utc).strftime('%Y-%m-%d %H:%M UTC')

lines = []
lines.append('πŸ“Š **LLM Snare Arena η›‘ζŽ§**')
lines.append(f'πŸ“… {now}')
lines.append('')
lines.append('**πŸ”₯ Top 5οΌˆθΏ‘ζœŸε‡εˆ†οΌ‰οΌš**')
for i,r in enumerate(rows[:5],1):
    lines.append(f'{i}. {r["name"]} β€” {r["avg8"]}')
lines.append('')
lines.append('**πŸ’€ εž«εΊ• 3：**')
for r in rows[-3:]:
    lines.append(f'- {r["name"]} β€” {r["avg8"]}')
lines.append('')
lines.append('**πŸ“ˆ θΏ‘ζœŸε˜εŒ–οΌˆε‰8 vs 再前8οΌ‰οΌš**')
changes = sorted(rows, key=lambda x: -abs(x['change']))[:6]
for r in changes:
    arrow = '↑' if r['change']>0 else '↓'
    lines.append(f'- {r["name"]} {arrow} {abs(r["change"])}')
lines.append('')
lines.append('**πŸ• ζœ€ζ–°δΈ€ζ¬‘εΎ—εˆ†οΌš**')
for r in rows:
    icon = 'βœ…' if r['latest'] >= 50 else '❌'
    lines.append(f'{icon} {r["name"]} β€” {r["latest"]} ({r["raw"]}/{r["max"]})')

result = '\n'.join(lines)
print(result)