Sari la conținut

Română · beta. Traducerea a fost făcută automat și este încă în curs de revizuire. Citește pagina în engleză

Benchmark-uri

Surse
8
Intrări urmărite
22.308
Ultima sincronizare
16 min
Toate sursele se sincronizează normal
  • Arena · Textacum 11 ore
  • Arena · WebDevacum 11 ore
  • Arena · Visionacum 11 ore
  • Arena · Agentacum 11 ore
  • GPQA Diamondacum 16 minute
  • SWE-benchacum 16 minute
  • Terminal-Benchacum 16 minute
  • ARC-AGI-2acum 16 minute

Gemini 3 Pro

1,521±22

1,499CI1,543

googleian. 2026oscilație #23–#90

Muse Spark

1,511±40

1,472CI1,551

metasept. 2026oscilație #16–#40

Germană ×1296 din 296 modele
ModelOrganizație± CIOscilațieAdaugă la comparație
1Muse Spark 1.3 (Max)metasept. 2026$1.25 / $4.251,5251,489–1,562#13–33
2Gemini 3 Progoogleian. 2026—1,5211,499–1,543#23–90
3Muse Sparkmetasept. 2026—1,5111,472–1,551#16–40
4Claude Opus 5 (Max)anthropicsept. 2026$5 / $251,5081,481–1,534#5–11
5Claude Opus 4.6 (High)anthropicsept. 2026$5 / $251,5071,490–1,524#2–25
6glm-5.3-maxzaiaug. 2026$0.08 / $1.401,5061,476–1,536#18–52
7Claude Opus 4.7 (High)anthropicsept. 2026$5 / $251,5051,486–1,524#1–18
8Kimi K3 (Max)moonshotiun. 2026$0.30 / $11.941,5051,481–1,529#10–39
9gpt-5.6-sol-xhighopenaisept. 2026$4 / $201,5041,481–1,528#6–25
10Claude Fable 5 (High)anthropicsept. 2026$10 / $501,5021,478–1,526#4–20
11deepseek-v4.1-flash-maxdeepseekoct. 2026$0.15 / $0.601,5001,463–1,537#23–31
12gemini-3-flashgoogleian. 2026—1,4991,474–1,525#38–153
13Gemini 3.7 Flash (High)googleoct. 2026$0.75 / $3.751,4991,470–1,528#9–57
14Gemini 3.1 Pro Previewgooglesept. 2026$2 / $121,4951,480–1,509#9–36
15claude-opus-4-8-highanthropicsept. 2026$5 / $251,4921,474–1,511#15–33
16Claude Fable 5.1 (Max)anthropicsept. 2026$10 / $501,4911,456–1,526#2–21
17gemini-3.5-flash-mediumgooglesept. 2026$1.50 / $91,4911,468–1,513#10–26
18Gemini 3.8 Flash (High)googleoct. 2026$0.83 / $4.131,4901,466–1,514#2–29
19gpt-5.5openaisept. 2026$5 / $301,4901,471–1,509#9–34
20gpt-6-astra-maxopenaisept. 2026$10 / $501,4901,447–1,533#1–18
21grok-4.20-beta1xaioct. 2026—1,4891,462–1,517#46–119
22qwen3.8-maxalibabaoct. 2026$2 / $61,4871,460–1,514#12–40
23gemini-3.6-flash-highgooglesept. 2026$0.75 / $3.751,4841,460–1,508#16–95
24gpt-5.2-chat-latest-20260210openaifeb. 2026—1,4811,455–1,506#44–72
25Muse Spark 1.1metasept. 2026$1.25 / $4.251,4801,457–1,502#21–115
26claude-opus-4-5-20251101anthropicnov. 2025$5 / $251,4781,460–1,497#20–62
27glm-5.2-maxzaiiun. 2026$0.62 / $2.421,4781,457–1,498#9–49
28gpt-4.5-preview-2025-02-27openaifeb. 2025—1,4771,445–1,509#74–530
29claude-opus-4-5-20251101-high-32kanthropicnov. 2025—1,4771,454–1,500#36–94
30gpt-5.6-luna-xhighopenaisept. 2026$0.20 / $1.201,4771,454–1,500#30–146
31gpt-5.6-terra-xhighopenaisept. 2026$2 / $121,4761,453–1,499#13–59
32gpt-5.4-highopenaisept. 2026$2.50 / $151,4761,457–1,495#12–29
33grok-4.20-multi-agent-beta-0309xaioct. 2026—1,4751,455–1,494#48–86
34glm-5.3-flashzaiaug. 2026$0.05 / $0.201,4741,448–1,499#26–60
35ernie-5.1baiduoct. 2026—1,4721,450–1,494#47–178

Se afișează Germană exact așa cum a publicat-o sursa — propriul scor și interval de încredere, fără mix. Selectează o a doua axă ca să construiești un scor compus ponderat.

Puținele lucruri din AI și cuantic care au contat cu adevărat săptămâna asta — rezumate cinstit.

Un e-mail pe săptămână. Dezabonare dintr-un clic. Fără urmărire, fără partajare.