Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d271Policy30d67, 15 pendingSources1,014
Pro Mode — raw scores, one benchmark under the lensofficial card harness
The field · 23 measured
41.480.6

23 models measured, most on the official card harness. Claude Opus 4.7 tops the board at 80.6.

23 measured·1 new this week·Lifecycle Updated Oct 8, 2026Score basis
#◆Vendor$/M in$/M out
Leader — ignites whiteField — fill cools with scoreBars show 41.4–80.6 · ◆ solid = first-party or better · outlined = arm's-length
01Claude Opus 4.7+2 altsAnthropic80.6$5.0/M$25.0/M$0.186
02Gemini 3.1 Pro+3 altsGoogle72.5$2.0/M$12.0/M$0.097
03Seed 2.1 Pro Preview+1 altByteDance72.2———
04Claude Fable 5.1+1 altAnthropic69$10.0/M$50.0/M$0.435
05Claude Opus 5.5+2 altsAnthropic67.7$4.0/M$20.0/M$0.177
06Claude Mythos 5+1 altAnthropic67.1$10.0/M$50.0/M$0.447
07Claude Opus 5+2 altsAnthropic66.9$5.0/M$25.0/M$0.224
08Hy4Tencent66.2$0.83/M$2.5/M$0.025
09Claude Opus 4.8+3 altsAnthropic66.2$5.0/M$25.0/M$0.227
10Sonnet 5.5+1 altAnthropic65.6$2.0/M$10.0/M$0.091
11Kimi K3+1 altMoonshot63.3$0.58/M$12.3/M$0.102
12GPT-5.6 SolOpenAI63.23P$4.0/M$20.0/M$0.190
13Seed2.1ByteDance62.8———
14GLM 5.3 FlashZ.ai62.4$0.15/M$0.50/M$0.005
15Haiku 5.5NEWAnthropic60.3$0.10/M$0.50/M$0.005
16Claude Sonnet 5+2 altsAnthropic59.4$2.0/M$10.0/M$0.101
17Claude Fable 5+2 altsAnthropic57.9$10.0/M$50.0/M$0.518
18Claude Opus 4.6Anthropic57.1$5.0/M$25.0/M$0.263
19GPT-5.5+3 altsOpenAI54.1$5.0/M$30.0/M$0.323
20Claude Sonnet 4.6Anthropic53.4$3.0/M$15.0/M$0.169
21GPT-5.4OpenAI51.1$2.5/M$15.0/M$0.171
22MiniMax M3MiniMax45.1$0.28/M$1.1/M$0.015
23GLM-5.2Z.ai41.43P$1.4/M$4.4/M$0.070
◆ solid = independently verified · aggregator attested · vendor attributed — outlined = cross-reference · source-attributed. Dimmed ◆ = confidence inferred. “+N alts” = alternate disclosures in the drawer with why-demoted. Prices per million tokens; $/point = input price ÷ score. Every row opens its model dossier.

Verification: 23 models scored · 0 independently verified · 4 vendor cross-reference · 19 vendor-reported · 1 with source disagreement. How these tiers are assigned