Skip to content
VECTOR WIREAI INTELLIGENCE
UTC
Pro Mode — raw scores, one benchmark under the lensthird-party eval harness
The field · 17 measured
24.2547.09

17 models measured, most on the third-party eval harness. Yi 1.5 34B 32K tops the board at 47.09.

17 measured·Lifecycle Updated Oct 8, 2026Score basis
#◆Vendor$/M in$/M out
Leader — ignites whiteField — fill cools with scoreBars show 24.25–47.09 · ◆ solid = first-party or better · outlined = arm's-length
01Yi 1.5 34B 32K01.AI47.09———
02Yi 1.5 34B01.AI46.66———
03Yi 1.5 34B Chat 16K01.AI45.45———
04Yi 1.5 34B Chat01.AI45.2———
05Marco-o1AIDC-AI41.17———
06Yi 1.5 9B Chat 16K01.AI39.94———
07Yi 1.5 9B Chat01.AI39.75———
08Yi 1.5 9B01.AI39.16———
09Yi 1.5 9B 32K01.AI37.65———
10Yi 9B 200K01.AI36.22———
11Yi 9B01.AI35.74———
12Yi 1.5 6B Chat01.AI31.93———
13Yi 1.5 6B01.AI31.44———
14NuminaMath-7B-CoTAI-MO28.68———
15NuminaMath-7B-TIRAI-MO27.33———
16Llama 3 Instruct 8BMeta25.97$0.04/M$0.14/M$0.004
17Yi Coder 9B Chat01.AI24.25———
◆ 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: 17 models scored · 17 independently verified · 0 with source disagreement. How these tiers are assigned