Skip to content
VECTOR WIREAI INTELLIGENCE
UTC
Pro Mode — raw scores, one benchmark under the lensthird-party eval harness
The field · 7 measured
82.989.1

7 models measured, most on the third-party eval harness. DeepSeek-V4-Flash tops the board at 89.1.

7 measured·Lifecycle Updated Oct 8, 2026Score basis
#◆Vendor$/M in$/M out
Leader — ignites whiteField — fill cools with scoreBars show 82.9–89.1 · ◆ solid = first-party or better · outlined = arm's-length
01DeepSeek-V4-FlashDeepSeek89.1$0.44/M$1.3/M$0.010
02DeepSeek-V4-ProDeepSeek88.9$0.43/M$0.87/M$0.007
03Qwen3.5 397B A17BAlibaba88.5$0.60/M$3.6/M$0.024
04Nemotron 3 Ultra 550B A55B+1 altNVIDIA88.4VENDOR$0.60/M$2.5/M$0.018
05MiniMax M2.7MiniMax84.9$0.21/M$0.84/M$0.006
06GLM-5.1Z.ai84.1$1.4/M$4.4/M$0.034
07Kimi K2.6Moonshot82.9$0.95/M$4.0/M$0.030
◆ 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: 7 models scored · 0 independently verified · 6 vendor cross-reference · 1 vendor-reported · 0 with source disagreement. How these tiers are assigned