Agentic Index benchmark

111 AI models ranked on Agentic Index. Each row shows the model’s score next to its cheapest API price per 1M tokens where it has one, so you can weigh quality against cost.

Top model today

GPT-5.6 Sol (OpenAI) leads at 54.0, from $5.00 per 1M input.

Data via Artificial Analysis

Models ranked

111

on this benchmark

Top score

54.0

current leader

Type

Composite

index

Updated

15 July 2026

last source fetch

Cost vs Agentic Index

71 priced models
Most attractive quadrant (cheap + high)
= reasoning model (extended thinking)

Each labelled name links to that model’s page. Curated to the cost-efficiency frontier, top scorers and cheapest so labels stay readable; the full field is in the leaderboard table below (and via “select all”). Reasoning models (extended “thinking” before answering) carry a ring around the dot.

Models ranked on Agentic Index

highest score first · 111 models
Maker
#
1GPT-5.6 SolOpenAI54.0$5.00$30.00
2Claude Fable 5Anthropic52.8$10.00$50.00
3GPT-5.6 TerraOpenAI47.4$2.50$15.00
4Claude Opus 4.8Anthropic47.2$5.00$25.00
5Claude Sonnet 5Anthropic46.7$2.00$10.00
6Grok 4.5xAI45.7$2.00$6.00
7GPT-5.6 LunaOpenAI45.6$1.00$6.00
8GPT-5.5OpenAI44.9$5.00$30.00
9Claude Opus 4.7Anthropic44.4$5.00$25.00
10GLM 5.2Zhipu43.1$1.40$4.40
11GPT-5.4OpenAI41.1$2.50$15.00
12Claude Sonnet 4.6Anthropic40.8reasoning: adaptive$3.00$15.00
13Muse Spark 1.1Other37.5$1.25$4.25
14Gemini 3.5 FlashGoogle37.4$1.50$9.00
15DeepSeek V4 ProDeepSeek36.4$0.43$0.87
16MiniMax M3MiniMax35.4$0.30$1.20
17JT-4.1 Flash 236B A21BChina-mobile34.9n/an/a
18DeepSeek V4 FlashDeepSeek31.1n/an/a
19Nex-N2-ProNex-agi31.0n/an/a
20Hy3-previewTencent30.7n/an/a
21Qwen3.7 MaxAlibaba30.6n/an/a
22Kimi K2.6Moonshot30.3n/an/a
23GPT-5.4 MiniOpenAI30.2$0.75$4.50
24GLM 5.1Zhipu29.9$1.05$3.50
25Kimi K2.7 CodeMoonshot29.6$0.95$4.00
26MiMo-V2.5-ProXiaomi29.1$1.00$3.00
27Muse SparkMeta28.7n/an/a
28Grok Build 0.1 0616Spacexai28.0n/an/a
29Qwen3.6 PlusAlibaba27.6$0.33$1.95
30GPT-5.4 NanoOpenAI27.5$0.20$1.25
31Nemotron 3 UltraNVIDIA27.4n/an/a
32Qwen3.6 27BAlibaba27.0$0.15$0.50
33GPT-5OpenAI25.7$1.25$10.00
34MiniMax M2.7MiniMax25.6$0.26$0.55
35GLM 4.7Zhipu25.4$0.40$1.50
36Claude 4.5 SonnetAnthropic24.6reasoning: true$3.00$15.00
37Grok 4.3xAI24.1$1.25$2.50
38MiMo-V2.5Xiaomi23.7n/an/a
39LongCat 2.0Longcat21.8n/an/a
40Kimi K2.5Moonshot21.7$0.50$2.80
41Step 3.7 FlashStepfun21.5n/an/a
42Qwen3.6 35B A3BAlibaba21.4$0.14$0.45
43Gemini 3.1 Pro PreviewGoogle21.4$2.00$12.00
44GPT-5.1OpenAI21.0$1.25$10.00
45Qwen3.7 PlusAlibaba20.8n/an/a
46Qwen3.5-122B-A10BAlibaba20.7$0.25$1.75
47Qwen3.5 397B A17BAlibaba19.8$0.60$3.60
48GPT-5 MiniOpenAI19.4$0.25$2.00
49Mistral Medium 3.5Mistral19.0$1.50$7.50
50Ring-2.6-1TInclusionai18.9n/an/a
150 of 111
Page 1 of 3
Prices are on-demand list rates. Enterprise commitments can be materially lower. Cost per task uses best-effort token usage.

Frequently asked questions

What is the Agentic Index benchmark?

Artificial Analysis's composite agentic score (0 to 100) measuring tool use, planning, autonomy and multi-step problem solving, computed by averaging its underlying agentic evaluations and normalising to a 0 to 100 scale.

Which AI model scores highest on Agentic Index?

GPT-5.6 Sol (OpenAI) leads with 54.0, from $5.00 per 1M input tokens.

How many models are ranked on Agentic Index?

111 models carry a Agentic Index score. Scores come from Artificial Analysis; where a model has a live API price it is verified against LiteLLM and OpenRouter, and models with no current price are listed without one.

Other benchmarks

compare the same models on a different eval
Benchmark scores via Artificial Analysis · pricing verified against LiteLLM + OpenRouter · refreshed daily · updated 15 July 2026

Every weekday

AI moves fast. Here's your debrief.

News, analysis, tools, and more.

For people who build with AI