Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d271Policy30d67, 15 pendingSources1,014
Pro Mode — raw scores, one benchmark under the lensAA-graded harness
The field · 43 measured
5.6556.215

43 models measured, most on the AA-graded harness. GPT-5.6 Sol tops the board at 56.215.

43 measured·4 new this week·Lifecycle Updated Oct 8, 2026Score basis
#◆Vendor$/M in$/M out
Leader — ignites whiteField — fill cools with scoreBars show 5.65–56.215 · ◆ solid = first-party or better · outlined = arm's-length
01GPT-5.6 SolOpenAI56.215$4.0/M$20.0/M$0.213
02Step 5 PreviewStepFun55.556$1.0/M$2.7/M$0.033
03Gemini 3.8 FlashGoogle52.542$0.75/M$3.8/M$0.043
04GLM 5.3 FlashZ.ai51.243$0.15/M$0.50/M$0.006
05GPT-5.6 TerraOpenAI51.036$2.0/M$12.0/M$0.137
06Claude Fable 5.1NEWAnthropic49.529$10.0/M$50.0/M$0.606
07GPT-6 SolOpenAI49.435$2.0/M$10.0/M$0.121
08GPT-6 AstraOpenAI48.588$10.0/M$50.0/M$0.617
09Kimi K3+1 altMoonshot47.693$0.58/M$12.3/M$0.135
10DeepSeek-V4.1-FlashDeepSeek46.916$0.20/M$0.60/M$0.009
11Claude Opus 4.7NEWAnthropic46.657$5.0/M$25.0/M$0.321
12GLM-5.3Z.ai46.083$1.4/M$4.4/M$0.063
13GPT-5.5OpenAI45.81$5.0/M$30.0/M$0.382
14GLM-5.2Z.ai42.655$1.4/M$4.4/M$0.068
15Qwen3.7 MaxAlibaba42.467$2.5/M$7.5/M$0.118
16Grok 4.7SpaceXAI42.081$2.0/M$6.0/M$0.095
17Gemini 3.5 FlashGoogle40.348$1.5/M$9.0/M$0.130
18GPT-5.6 LunaOpenAI40.32$0.20/M$1.2/M$0.017
19Qwen3.8 Max PreviewAlibaba40.254$3.3/M$9.9/M$0.164
20GLM-5.1Z.ai40.254$1.4/M$4.4/M$0.072
21Claude Sonnet 4.6NEWAnthropic39.802$3.0/M$15.0/M$0.226
22DeepSeek-V4-Pro+1 altDeepSeek38.324$0.43/M$0.87/M$0.017
23Claude Opus 5.5NEWAnthropic38.23$4.0/M$20.0/M$0.314
24MiMo V2.5 ProXiaomi38.23$0.43/M$0.87/M$0.017
25Gemma 4 31BGoogle37.288$0.17/M$0.40/M$0.008
26Qwen3.5 27BAlibaba35.499$0.30/M$2.4/M$0.038
27GPT-5.4 mini+1 altOpenAI35.217$0.75/M$4.5/M$0.075
28GPT-5.4+1 altOpenAI34.5$2.5/M$15.0/M$0.254
29Qwen3.5 397B A17BAlibaba34.087$0.60/M$3.6/M$0.062
30Muse Spark 1.3Meta33.239$1.3/M$4.3/M$0.083
31Grok 4.3SpaceXAI32.721$1.3/M$2.5/M$0.057
32DeepSeek-V4-Flash+1 altDeepSeek31.516$0.44/M$1.3/M$0.028
33Kimi K2.6Moonshot31.186$0.95/M$4.0/M$0.079
34Gemini 3.1 ProGoogle30.331$2.0/M$12.0/M$0.231
35Step 3.7 FlashStepFun30.273$0.20/M$1.1/M$0.022
36Claude Haiku 4.5Anthropic27.307$1.0/M$5.0/M$0.110
37MiniMax M2.7MiniMax26.46$0.21/M$0.84/M$0.020
38GPT-5.4 nano+1 altOpenAI24.388$0.20/M$1.3/M$0.030
39Gemma 4 26B A4BGoogle23.635$0.07/M$0.34/M$0.009
40Qwen3.5 35B A3BAlibaba21.516$0.25/M$2.0/M$0.052
41Grok 4.1 FastSpaceXAI17.891$0.20/M$0.50/M$0.020
42Muse SparkMeta16.205———
43GPT Oss 120bOpenAI5.65$0.15/M$0.59/M$0.066
◆ 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: 43 models scored · 0 independently verified · 43 aggregator-attributed · 0 with source disagreement. How these tiers are assigned