Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d263Policy30d65, 14 pendingSources1,013
Pro Mode — raw scores, one benchmark under the lensthird-party eval harness
The field · 101 measured
76.598.2

101 models measured, most on the third-party eval harness. Finix_s1_32b tops the board at 98.2.

101 measured·Lifecycle Updated Oct 8, 2026Score basis
#◆Vendor$/M in$/M out
Leader — ignites whiteField — fill cools with scoreBars show 76.5–98.2 · ◆ solid = first-party or better · outlined = arm's-length
01Finix_s1_32bantgroup98.2———
02GPT-5.4 nanoOpenAI96.9$0.20/M$1.3/M$0.007
03Gemini 2.5 Flash LiteGoogle96.7$0.10/M$0.40/M$0.003
04Phi 4Microsoft96.3VENDOR_EVA$0.13/M$0.50/M$0.003
05Llama 3.3 70B InstructMeta95.9$0.71/M$0.72/M$0.007
06Arctic InstructSnowflake95.7———
07Gemma 3 12BGoogle95.6$0.05/M$0.15/M$0.001
08Mistral Large 2.1Mistral95.5VENDOR_EVA$2.0/M$6.0/M$0.042
09Qwen3 8BAlibaba95.2$0.18/M$0.70/M$0.005
10Nova 2.0 LiteAmazon94.9$0.30/M$2.5/M$0.015
11Nova ProAmazon94.9$0.80/M$3.2/M$0.021
12Mistral Small 3 (Non-Reasoning)Mistral94.9$0.05/M$0.08/M$0.001
13Granite 4.0 H Small+1 altIBM94.8$0.06/M$0.25/M$0.002
14Gemma 4 26B A4BGoogle94.8$0.07/M$0.34/M$0.002
15Jamba Mini 2AI2194.7———
16DeepSeek-V3.2-ExpDeepSeek94.7$0.28/M$0.42/M$0.004
17Qwen3 14BAlibaba94.6$0.35/M$1.4/M$0.009
18GPT-5.4 miniOpenAI94.5$0.75/M$4.5/M$0.028
19DeepSeek-V3.1DeepSeek94.5$0.56/M$1.7/M$0.012
20Nova MicroAmazon94.5$0.04/M$0.14/M$0.001
21GPT-4.1OpenAI94.4$2.0/M$8.0/M$0.053
22Qwen3 4BAlibaba94.3———
23Grok 3+1 altSpaceXAI94.2$4.0/M$20.0/M$0.127
24Qwen3 32BAlibaba94.1$0.16/M$0.64/M$0.004
25DeepSeek-V3DeepSeek93.9$0.24/M$0.90/M$0.006
26Nova LiteAmazon93.9$0.06/M$0.24/M$0.002
27DeepSeek-V3.2DeepSeek93.7$0.28/M$0.42/M$0.004
28Gemma 3 4BGoogle93.6$0.05/M$0.10/M$0.001
29GPT-6 SolOpenAI93.5VENDOR_EVA$2.0/M$10.0/M$0.064
30Trinity Large ThinkingArcee AI93.1$0.25/M$0.80/M$0.006
31Command R+ (Apr '24)Cohere93.1$2.5/M$10.0/M$0.067
32Gemini 2.5 ProGoogle93$1.3/M$10.0/M$0.060
33GPT-5.4OpenAI93$2.5/M$15.0/M$0.094
34Ministral 3B 2410Mistral92.7———
35Gemma 3 27BGoogle92.6$0.08/M$0.16/M$0.001
36Ministral 8B 2410Mistral92.6———
37Gemma 4 31BGoogle92.6$0.17/M$0.40/M$0.003
38Llama 4 ScoutMeta92.3$0.19/M$0.68/M$0.005
39Gemini 2.5 FlashGoogle92.2$0.30/M$2.5/M$0.015
40Gemini 3.1 Flash Lite PreviewGoogle91.8$0.25/M$1.5/M$0.010
41Llama 4 Maverick 17B 128E Instruct FP8Meta91.8$0.27/M$0.85/M$0.006
42GPT-5.4 ProOpenAI91.7$30.0/M$180.0/M$1.145
43DeepSeek-V4-ProDeepSeek91.4$0.43/M$0.87/M$0.007
44GPT-6 AstraOpenAI91.3VENDOR_EVA$10.0/M$50.0/M$0.329
45MiniMax M2.5MiniMax90.9$0.27/M$1.1/M$0.007
46GLM-4.5-AirZ.ai90.7$0.17/M$0.98/M$0.006
47GLM-4.7-FlashZ.ai90.7$0.06/M$0.40/M$0.003
48Command ACohere90.7$2.5/M$10.0/M$0.069
49Qwen3 235B A22BAlibaba90.7$0.70/M$2.8/M$0.019
50GPT-5.5OpenAI90.7$5.0/M$30.0/M$0.193
51Qwen3 Next 80B A3B+1 altAlibaba90.7$0.15/M$1.2/M$0.007
52GLM-4.6Z.ai90.5$0.57/M$2.2/M$0.015
53Aya Expanse 8BCohere90.5———
54NVIDIA Nemotron 3 Nano 30B A3B+1 altNVIDIA90.4VENDOR_EVA$0.05/M$0.20/M$0.001
55GPT-4oOpenAI90.4$5.0/M$15.0/M$0.111
56Jamba 1.7 LargeAI2190.3———
57Claude Haiku 4.5Anthropic90.2$1.0/M$5.0/M$0.033
58GLM-5Z.ai89.9$1.0/M$3.2/M$0.023
59Claude Sonnet 4+1 altAnthropic89.7$3.0/M$15.0/M$0.100
60Gemini 3.1 ProGoogle89.6$2.0/M$12.0/M$0.078
61GPT-5 nanoOpenAI89.5$0.05/M$0.40/M$0.003
62Qwen3.5 FlashAlibaba89.5$0.10/M$0.40/M$0.003
63Qwen3.5 35B A3B+1 altAlibaba89.5$0.25/M$2.0/M$0.013
64Granite 3.3 8B InstructIBM89.4$0.03/M$0.25/M$0.002
65Claude Sonnet 4.6Anthropic89.4$3.0/M$15.0/M$0.101
66Qwen3.5 Plus 02.15Alibaba89.3$0.26/M$1.6/M$0.010
67Kimi K2.6Moonshot89.2$0.95/M$4.0/M$0.028
68GPT-5.2+2 altsOpenAI89.2$1.8/M$14.0/M$0.088
69Claude Opus 4.5+1 altAnthropic89.1$5.0/M$25.0/M$0.168
70Aya Expanse 32BCohere89.1———
71Qwen3.5 122B A10B+1 altAlibaba88.8$0.40/M$3.2/M$0.020
72DeepSeek-R1+1 altDeepSeek88.7$2.0/M$4.0/M$0.034
73GLM-4.7Z.ai88.3VENDOR_EVA$0.60/M$2.2/M$0.016
74Claude Opus 4.1Anthropic88.2$15.0/M$75.0/M$0.510
75MiniMax M2.1MiniMax88.2$0.30/M$1.2/M$0.009
76Claude Sonnet 4.5+1 altAnthropic88$3.0/M$15.0/M$0.102
77Claude Opus 4.7+1 altAnthropic88$5.0/M$25.0/M$0.170
78Claude Opus 4+1 altAnthropic88VENDOR_EVA$15.0/M$75.0/M$0.511
79GPT-5.1+3 altsOpenAI87.9$1.3/M$10.0/M$0.064
80Qwen3.5 27BAlibaba87.9$0.30/M$2.4/M$0.015
81Claude Opus 4.6Anthropic87.8$5.0/M$25.0/M$0.171
82Mercury 2Inception87.7$0.25/M$0.75/M$0.006
83GPT-5.6 SolOpenAI87.6VENDOR_EVA$4.0/M$20.0/M$0.137
84GPT-5 miniOpenAI87.1$0.25/M$2.0/M$0.013
85MiniMax M2.7MiniMax87.1VENDOR_EVA$0.21/M$0.84/M$0.006
86Gemini 3 Flash PreviewGoogle86.5$0.50/M$3.0/M$0.020
87Gemini 3 ProGoogle86.4$2.0/M$12.0/M$0.081
88GPT Oss 120bOpenAI85.8$0.15/M$0.59/M$0.004
89Kimi K2.5Moonshot85.8$0.45/M$2.3/M$0.016
90Mistral Large 3Mistral85.5$0.50/M$1.5/M$0.012
91Jamba 1.7 MiniAI2185.3———
92GPT-5+3 altsOpenAI84.9$1.3/M$10.0/M$0.066
93Kimi K2 InstructMoonshot82.1$0.57/M$2.3/M$0.017
94O4 Mini+2 altsOpenAI81.4$1.1/M$4.4/M$0.034
95Grok 4.1 Fast+2 altsSpaceXAI80.8$0.20/M$0.50/M$0.004
96Ministral 3 14BMistral80.6$0.20/M$0.20/M$0.002
97Grok 4 Fast+2 altsSpaceXAI79.8$0.20/M$0.50/M$0.004
98Ministral 3 8BMistral78.3$0.15/M$0.15/M$0.002
99Mistral Medium 3.1 (Non-Reasoning)+1 altMistral77.3$0.40/M$2.0/M$0.016
100O3 Pro+1 altOpenAI76.7$20.0/M$80.0/M$0.652
101Phi 4 Mini Instruct+1 altMicrosoft76.5———
◆ 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: 101 models scored · 101 independently verified · 0 with source disagreement. How these tiers are assigned