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
22.318
Ultima sincronizare
20 min
Toate sursele se sincronizează normal
  • Arena · Textacum 5 ore
  • Arena · WebDevacum 5 ore
  • Arena · Visionacum 5 ore
  • Arena · Agentacum 5 ore
  • GPQA Diamondacum 20 de minute
  • SWE-benchacum 20 de minute
  • Terminal-Benchacum 20 de minute
  • ARC-AGI-2acum 20 de minute
Coreeană ×1265 din 265 modele
ModelOrganizație± CIOscilațieAdaugă la comparație
1Claude Fable 5.1 (Max)anthropicsept. 2026$10 / $501,5011,463–1,540#2–22
2Claude Fable 5 (High)anthropicsept. 2026$10 / $501,4991,477–1,522#4–21
3Claude Opus 5 (Max)anthropicsept. 2026$5 / $251,4981,474–1,523#5–11
4Muse Spark 1.3 (Max)metasept. 2026$1.25 / $4.251,4951,463–1,528#13–33
5Muse Spark 1.2 (xHigh)metaoct. 2026$1.25 / $4.251,4811,439–1,522#13–112
6Gemini 3.7 Flash (High)googleoct. 2026$0.75 / $3.751,4771,449–1,505#9–57
7Muse Spark 1.1metasept. 2026$1.25 / $4.251,4741,453–1,496#21–115
8Muse Sparkmetasept. 2026—1,4741,435–1,513#16–41
9glm-5.3-maxzaiaug. 2026$0.08 / $1.401,4721,442–1,502#18–53
10Kimi K3 (Max)moonshotiun. 2026$0.24 / $11.941,4721,449–1,494#10–40
11Claude Opus 4.7anthropicsept. 2026$5 / $251,4711,453–1,490#1–18
12deepseek-v4-pro-high-20260813deepseekaug. 2026$0.66 / $1.981,4651,425–1,506#17–55
13Claude Opus 4.6anthropicsept. 2026$5 / $251,4651,448–1,482#2–25
14qwen3.8-maxalibabaoct. 2026$2 / $61,4581,433–1,484#12–41
15gpt-5.6-sol-xhighopenaisept. 2026$4 / $201,4561,433–1,478#6–24
16Gemini 3.1 Pro Previewgooglesept. 2026$2 / $121,4541,440–1,468#9–36
17gpt-6-astra-maxopenaisept. 2026$10 / $501,4531,413–1,494#1–18
18deepseek-v4.1-flash-maxdeepseekoct. 2026$0.30 / $1.201,4531,414–1,492#14–30
19gpt-5.5-instantopenaisept. 2026—1,4511,421–1,481#52–114
20Gemini 3.8 Flash (High)googleoct. 2026$0.83 / $4.131,4501,425–1,475#2–30
21claude-opus-4-8-highanthropicsept. 2026$5 / $251,4491,430–1,469#15–33
22Gemini 3 Progoogleian. 2026—1,4481,426–1,471#23–91
23claude-opus-4-5-20251101anthropicnov. 2025$5 / $251,4471,429–1,466#19–62
24gemini-3-flashgoogleian. 2026—1,4451,419–1,470#38–153
25grok-4.20-beta1xaioct. 2026—1,4451,417–1,473#46–119
26gpt-5.5-highopenaisept. 2026$5 / $301,4431,425–1,462#9–34
27glm-5.2-maxzaiiun. 2026$0.62 / $2.421,4431,423–1,463#9–49
28gemini-3.5-flash-mediumgooglesept. 2026$1.50 / $91,4401,419–1,462#10–26
29glm-5.3-flashzaiaug. 2026$0.04 / $0.201,4401,414–1,465#26–59
30mimo-v2.5-proxiaomiapr. 2026$0.43 / $0.871,4361,419–1,454#40–129
31deepseek-v4-pro-high-previewdeepseekoct. 2026—1,4361,416–1,456#63–175
32claude-sonnet-5-highanthropicsept. 2026$2 / $101,4341,413–1,455#24–58
33grok-4.20-multi-agent-beta-0309xaioct. 2026—1,4341,414–1,453#48–86
34claude-sonnet-4-6anthropicsept. 2026$3 / $151,4331,414–1,451#39–78
35grok-4.7-xhighxaioct. 2026$2 / $61,4331,389–1,476#23–72

Se afișează Coreeană 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.