Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d261Policy30d64, 10 pendingSources1,015
Pro Mode — raw scores, one benchmark under the lensofficial card harness
The field · 39 measured
5.5691.9

39 models measured, most on the official card harness. Claude Opus 4.6 tops the board at 91.9.

39 measured·Lifecycle Updated Oct 9, 2026Score basis
#◆Vendor$/M in$/M out
Leader — ignites whiteField — fill cools with scoreBars show 5.56–91.9 · ◆ solid = first-party or better · outlined = arm's-length
01Claude Opus 4.6Anthropic91.9$5.0/M$25.0/M$0.163
02Claude Sonnet 4.6Anthropic91.7$3.0/M$15.0/M$0.098
03Claude Opus 4.5+1 altAnthropic88.9$5.0/M$25.0/M$0.169
04Claude Opus 4.1Anthropic86.8$15.0/M$75.0/M$0.518
05Claude Sonnet 4.5Anthropic86.2$3.0/M$15.0/M$0.104
06Gemini 3 ProGoogle85.3$2.0/M$12.0/M$0.082
07Claude Haiku 4.5Anthropic83.2$1.0/M$5.0/M$0.036
08GPT-5.2OpenAI82$1.8/M$14.0/M$0.096
09Claude Opus 4+1 altAnthropic81.83P$15.0/M$75.0/M$0.550
10GPT-5OpenAI81.1$1.3/M$10.0/M$0.069
11O3OpenAI80.2$2.0/M$8.0/M$0.062
12EXAONE 4.5 33BLG AI77.9———
13GPT-5.1OpenAI77.9$1.3/M$10.0/M$0.072
14GPT-5.1 InstantOpenAI77.9———
15Claude Sonnet 4+1 altAnthropic753P$3.0/M$15.0/M$0.120
16LongCat Flash LiteMeituan73.1———
17Qwen3.5 4BAlibaba71.933P$0.03/M$0.15/M$0.001
18Qwen3 235B A22B Instruct 2507Alibaba71.3$0.23/M$0.92/M$0.008
19Kimi K2 Instruct+3 altsMoonshot70.6$0.57/M$2.3/M$0.020
20DeepSeek-V3+1 altDeepSeek69.13P$0.24/M$0.90/M$0.008
21Qwen3 Next 80B A3BAlibaba67.8$0.15/M$1.2/M$0.010
22GPT-4oOpenAI63.4$5.0/M$15.0/M$0.158
23Nemotron 3 Super 120B A12BNVIDIA62.83$0.30/M$0.90/M$0.010
24Qwen3 Next 80B A3B InstructAlibaba57.3$0.15/M$1.2/M$0.012
25Qwen3 235B A22B+1 altAlibaba573P$0.70/M$2.8/M$0.031
26NVIDIA Nemotron 3 Nano 30B A3BNVIDIA56.9$0.05/M$0.20/M$0.002
27Qwen3 30B A3B 2507 ThinkingAlibaba56.143P$0.20/M$2.4/M$0.023
28Gemma 4 E4BGoogle42.113P$0.02/M$0.10/M$0.001
29LFM2.5-8B-A1BLiquid AI39.823P———
30Gemma 4 E2BGoogle18.953P———
31Granite 4.0 H TinyIBM18.423P———
32LFM2.5-350M+1 altLiquid AI17.843P———
33LFM2.5-230MLiquid AI13.683P———
34LFM2-8B-A1B (Non-Reasoning)Liquid AI7.023P———
35Qwen3.5 0.8BAlibaba7.023P———
36Gemma 3 1B Instruct+1 altGoogle6.433P———
37Qwen3.5 0.8B (Instruct)+1 altAlibaba6.143P———
38Granite 4.0 350M+1 altIBM6.143P———
39LFM2-350M+1 altLiquid AI5.563P———
◆ 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: 39 models scored · 0 independently verified · 16 vendor cross-reference · 23 vendor-reported · 0 with source disagreement. How these tiers are assigned