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
22.310
Ultima sincronizare
58 min
Toate sursele se sincronizează normal
  • Arena · Textacum 3 ore
  • Arena · WebDevacum 3 ore
  • Arena · Visionacum 3 ore
  • Arena · Agentacum 3 ore
  • GPQA Diamondacum 58 de minute
  • SWE-benchacum 58 de minute
  • Terminal-Benchacum 58 de minute
  • ARC-AGI-2acum 58 de minute
Conversații multi-tur ×1394 din 394 modele
ModelOrganizație± CIOscilațieAdaugă la comparație
1Gemini 4 Argon (High)googleoct. 2026$2 / $101,5521,529–1,575#1–10
2Muse Wonder (Max)metaoct. 2026—1,5291,501–1,558#14–56
3Claude Fable 5 (High)anthropicsept. 2026$10 / $501,5181,510–1,526#4–20
4Claude Opus 4.6 (High)anthropicsept. 2026$5 / $251,5171,511–1,523#2–25
5Claude Opus 4.7 (High)anthropicsept. 2026$5 / $251,5161,509–1,523#1–18
6Muse Spark 1.2 (xHigh)metaoct. 2026$1.25 / $4.251,5141,497–1,532#13–112
7Gemini 3.8 Flash (High)googleoct. 2026$0.83 / $4.131,5031,494–1,512#2–29
8Claude Opus 5.5 (High)anthropicoct. 2026$4 / $201,5011,482–1,519#2–11
9Kimi K3 (Max)moonshotiun. 2026$0.25 / $11.941,5001,491–1,509#10–39
10claude-opus-4-8-highanthropicsept. 2026$5 / $251,4961,490–1,503#15–33
11Gemini 3.1 Pro Previewgooglesept. 2026$2 / $121,4961,491–1,501#9–36
12Gemini 3 Progoogleian. 2026—1,4961,488–1,503#23–90
13Muse Spark 1.1metasept. 2026$1.25 / $4.251,4951,487–1,503#21–115
14gpt-5.2-chat-latest-20260210openaifeb. 2026—1,4951,487–1,502#44–72
15Muse Sparkmetasept. 2026—1,4941,481–1,507#16–40
16gpt-5.4-highopenaisept. 2026$2.50 / $151,4931,486–1,500#12–29
17Muse Spark 1.3 (Max)metasept. 2026$1.25 / $4.251,4911,479–1,503#13–33
18qwen3.8-maxalibabaoct. 2026$2 / $61,4901,481–1,500#12–40
19Gemini 3.7 Flash (High)googleoct. 2026$0.75 / $3.751,4901,480–1,499#9–57
20gpt-5.6-sol-xhighopenaisept. 2026$4 / $201,4881,480–1,496#6–25
21gemini-3.6-flash-highgooglesept. 2026$0.75 / $3.751,4871,479–1,496#16–95
22gpt-5.5-highopenaisept. 2026$5 / $301,4871,480–1,494#9–34
23Claude Fable 5.1 (Max)anthropicsept. 2026$10 / $501,4871,473–1,501#2–21
24gpt-6.1-sol-maxopenaioct. 2026$2 / $101,4861,465–1,507#1–15
25Claude Opus 5 (Max)anthropicsept. 2026$5 / $251,4861,477–1,495#5–11
26claude-opus-4-5-20251101-high-32kanthropicnov. 2025—1,4861,478–1,493#36–94
27gpt-6-astra-maxopenaisept. 2026$10 / $501,4861,471–1,500#1–18
28claude-opus-4-5-20251101anthropicnov. 2025$5 / $251,4841,478–1,490#20–62
29grok-4.20-beta1xaioct. 2026—1,4831,474–1,493#46–119
30qwen3.7-max-previewalibabaoct. 2026—1,4831,460–1,505#29–73
31gpt-5.5-instantopenaisept. 2026—1,4821,473–1,492#52–114
32gemini-3-flashgoogleian. 2026—1,4821,473–1,490#38–153
33mimo-v2.5-proxiaomiapr. 2026$0.43 / $0.871,4811,475–1,488#40–129
34claude-sonnet-4-6anthropicsept. 2026$3 / $151,4811,475–1,488#38–77
35gemini-3.5-flash-highgooglesept. 2026$1.50 / $91,4801,473–1,488#10–26

Se afișează Conversații multi-tur 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.