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 · 66 measured
22.582.7

66 models measured, most on the official card harness. GPT-5.5 tops the board at 82.7.

66 measured·Lifecycle Updated Oct 9, 2026Score basis
#◆Vendor$/M in$/M out
Leader — ignites whiteField — fill cools with scoreBars show 22.5–82.7 · ◆ solid = first-party or better · outlined = arm's-length
01GPT-5.5OpenAI82.7$5.0/M$30.0/M$0.212
02Claude Mythos PreviewAnthropic82———
03Claude Sonnet 5Anthropic80.4$2.0/M$10.0/M$0.075
04GPT-5.3 CodexOpenAI77.3$1.8/M$14.0/M$0.102
05Gemini 3.5 FlashGoogle76.2$1.5/M$9.0/M$0.069
06GPT-5.4+8 altsOpenAI75.1$2.5/M$15.0/M$0.117
07Claude Opus 4.8Anthropic74.6$5.0/M$25.0/M$0.201
08Qwen3.7 Plus PreviewAlibaba70.3$0.40/M$1.6/M$0.014
09Qwen3.7 MaxAlibaba69.7$2.5/M$7.5/M$0.072
10Claude Opus 4.7+2 altsAnthropic69.4$5.0/M$25.0/M$0.216
11GLM-5.1+4 altsZ.ai69$1.4/M$4.4/M$0.042
12Gemini 3.1 Pro+7 altsGoogle68.5$2.0/M$12.0/M$0.102
13MiMo V2.5 Pro+1 altXiaomi68.4$0.43/M$0.87/M$0.010
14DeepSeek-V4-Pro+7 altsDeepSeek67.9$0.43/M$0.87/M$0.010
15Kimi K2.6+3 altsMoonshot66.7$0.95/M$4.0/M$0.037
16MiMo V2.5Xiaomi65.8$0.14/M$0.28/M$0.003
17Claude Opus 4.6+6 altsAnthropic65.4$5.0/M$25.0/M$0.229
18GPT-5.2 CodexOpenAI64$1.8/M$14.0/M$0.123
19Qwen3.6 PlusAlibaba61.6$0.50/M$3.0/M$0.028
20GPT-5.1 Codex MaxOpenAI60.43P$1.3/M$10.0/M$0.093
21GPT-5.4 miniOpenAI60$0.75/M$4.5/M$0.044
22Qwen3.6 27B+1 altAlibaba59.3$0.60/M$3.6/M$0.035
23Claude Opus 4.5+4 altsAnthropic59.3$5.0/M$25.0/M$0.253
24Claude Sonnet 4.6Anthropic59.1$3.0/M$15.0/M$0.152
25Muse SparkMeta59———
26MiMo V2 ProXiaomi57.1———
27MiniMax M2.7+3 altsMiniMax57$0.21/M$0.84/M$0.009
28DeepSeek-V4-FlashDeepSeek56.9$0.44/M$1.3/M$0.015
29GLM-5+3 altsZ.ai56.2$1.0/M$3.2/M$0.037
30North-Mini-Code-1.0+1 altCohere55.1———
31MAI-Code-1-FlashMicrosoft54.8———
32Gemini 3 Pro+5 altsGoogle54.2$2.0/M$12.0/M$0.129
33GPT-5.2+1 altOpenAI543P$1.8/M$14.0/M$0.146
34GPT-5.1 Codex+1 altOpenAI52.8$1.3/M$10.0/M$0.107
35Qwen3.5 397B A17B+1 altAlibaba52.5$0.60/M$3.6/M$0.040
36Qwen3.6 35B A3B+2 altsAlibaba51.5$0.38/M$2.3/M$0.025
37Step 3.5 Flash+3 altsStepFun51$0.10/M$0.30/M$0.004
38Kimi K2.5+4 altsMoonshot50.8$0.45/M$2.3/M$0.027
39Claude Sonnet 4.5+3 altsAnthropic50$3.0/M$15.0/M$0.180
40Qwen3.5 122B A10BAlibaba49.4$0.40/M$3.2/M$0.036
41MiniMax M2.1+3 altsMiniMax47.93P$0.30/M$1.2/M$0.016
42Gemini 3 Flash PreviewGoogle47.6$0.50/M$3.0/M$0.037
43GPT-5.1+1 altOpenAI47.6$1.3/M$10.0/M$0.118
44Claude Opus 4.1Anthropic46.5$15.0/M$75.0/M$0.968
45DeepSeek-V3.2+11 altsDeepSeek46.4$0.28/M$0.42/M$0.008
46DeepSeek-V3.2-SpecialeDeepSeek46.4$0.29/M$0.43/M$0.008
47GPT-5.4 nanoOpenAI46.3$0.20/M$1.3/M$0.016
48MAI-Thinking-1+1 altMicrosoft46———
49Seed 1.8ByteDance45.2$0.25/M$2.0/M$0.025
50Gemma 4 31B+1 altGoogle42.9$0.17/M$0.40/M$0.007
51Qwen3.5 27B+2 altsAlibaba41.6$0.30/M$2.4/M$0.032
52GLM-4.7+4 altsZ.ai41$0.60/M$2.2/M$0.034
53Qwen3.5 35B A3B+1 altAlibaba40.5$0.25/M$2.0/M$0.028
54MiMo V2 Flash+3 altsXiaomi38.5$0.10/M$0.30/M$0.005
55Qwen3 Coder 480B A35B InstructAlibaba37.5$1.5/M$7.5/M$0.120
56Laguna-XS-2.1poolside37.5$0.06/M$0.12/M$0.002
57Kimi K2 Thinking+4 altsMoonshot35.7$0.60/M$2.5/M$0.043
58GPT-5OpenAI35.23P$1.3/M$10.0/M$0.160
59Gemma 4 26B A4BGoogle34.2$0.07/M$0.34/M$0.006
60Devstral 2Mistral32.63P$0.40/M$2.0/M$0.037
61Nemotron 3 Super 120B A12BNVIDIA31$0.30/M$0.90/M$0.019
62MiniMax M2+1 altMiniMax303P$0.30/M$1.2/M$0.025
63Qwen3 Coder PlusAlibaba25.43P$1.0/M$5.0/M$0.118
64GLM-4.6+1 altZ.ai24.53P$0.57/M$2.2/M$0.057
65Devstral Small 2Mistral22.53P———
66Qwen3 Max (Reasoning)Alibaba22.5$1.2/M$6.0/M$0.160
◆ 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: 66 models scored · 0 independently verified · 8 vendor cross-reference · 58 vendor-reported · 5 with source disagreement. How these tiers are assigned