Skip to content
Get my AI grade

Rank.ai Index

Original data on AI answers, GPU prices and the AI buildout

Rank.ai records what ChatGPT, Claude and Gemini recommend, logs what AI compute costs, and follows the data center buildout. Every finding below is dated, sourced and free to cite, with its own link, embed and PNG.

Latest snapshot Oct 16, 2026. Every board shows its own date.

544

Published boards

Every one is in the data ledger

29,480

Ranked rows

16

Kinds of thing ranked

6

Collections

AI models

The LLM leaderboard: every major AI benchmark, blended into one ranking

4 public leaderboards rank AI models their own way. We blend them into one consensus score, refreshed as they publish.

Best AI model right nowClaude Fable 5Anthropic · #1 on 1 of 3 leaderboards85/100 consensusBlended from Artificial Analysis, LiveBench, LMArena and Berkeley BFCL.
  1. 2GPT-5.6 SolOpenAI84
  2. 3Claude Opus 5Anthropic83
  3. 4GPT-5.5OpenAI82
  4. 5Kimi K3Moonshot AI82
  5. 6Claude Opus 4.8Anthropic81
  6. All 121 models across AA, LiveBench, Arena, BFCL →

Findings

What the latest data says

The lead finding from each data page. Rank.ai original data is marked; the rest names its source.

Rank.ai original data#

1 of 24answers naming Oracle Cloud Infrastructure

Oracle plans 8.3 GW of US AI data center capacity, more than any other cloud. ChatGPT, Claude and Gemini named Oracle Cloud Infrastructure in 1 of 24 GPU cloud recommendations.

Capacity by owner from Epoch AI, Oct 15, 2026. Recommendations from the 24 answers Rank.ai recorded to GPU cloud buying questions on Sep 17, 2026. The cloud named most was Runpod, in 14 of 24.

  1. OracleOracle Cloud Infrastructure8.3 GW1 of 24
  2. AmazonAmazon EC24.3 GW1 of 24
  3. GoogleGoogle Cloud4.1 GW10 of 24
  4. MicrosoftMicrosoft Azure3.7 GW9 of 24
  5. CoreWeave2.2 GW10 of 24

Owners are as Epoch AI labels them. A recommendation is an answer that names the cloud when asked which GPU cloud to use.

See the data center map

Source: Epoch AI, Data on AI data centers (CC BY 4.0); Rank.ai, Cloud GPU platform rank (recorded AI answers). Data as of Sep 17, 2026.

Rank.ai original data#

9 of 40questions with the same first business

ChatGPT, Claude and Gemini named the same local business first for 9 of 40 questions

40 local buying questions in Austin, TX, asked once of each engine on September 25, 2026.

  1. All three named the same business first9 of 40
  2. Two of the three agreed24 of 40
  3. Each named a different business first7 of 40
Read the local AI report

Source: Rank.ai, Local AI recommendations report, October 2026. Data as of Sep 25, 2026.

Epoch AI data, Rank.ai analysis#

42.5%of 33.7 GW in 3 states

Texas, Wisconsin and Indiana hold 42.5% of planned US AI data center capacity

33.7 GW is planned across 20 states at 39 reviewed sites, as of Oct 15, 2026.

  1. Texas8 sites6.9 GW
  2. Wisconsin2 sites4.0 GW
  3. Indiana2 sites3.3 GW
  4. New Mexico2 sites2.7 GW
  5. Ohio4 sites2.7 GW
  6. Louisiana1 site2.2 GW
  7. Tennessee1 site2.1 GW
  8. Iowa2 sites1.9 GW
  9. Mississippi2 sites1.8 GW
  10. Michigan1 site1.4 GW

Planned facility power is estimated capacity. Planned does not mean permitted, financed, built or switched on.

See the data center map

Source: Epoch AI, Data on AI data centers (CC BY 4.0); Rank.ai, Planned AI data center capacity by state. Data as of Oct 15, 2026.

Rank.ai original data#

6 vs 0CoreWeave, ChatGPT vs Claude

ChatGPT recommended CoreWeave in 6 of 8 GPU cloud answers. Claude recommended it in 0 of 8.

Claude never recommended CoreWeave, Google Cloud, Microsoft Azure or Hyperstack, each of which ChatGPT recommended at least once.

  1. Runpod7 of 82 of 8
  2. Lambda Cloud6 of 82 of 8
  3. CoreWeave6 of 80 of 8
  4. Google Cloud4 of 80 of 8
  5. Microsoft Azure4 of 80 of 8
  6. Vast.ai4 of 81 of 8
  7. Hyperstack1 of 80 of 8
  8. IBM Cloud1 of 80 of 8

ChatGPT and Claude through their providers' APIs, asked the same GPU cloud questions on the same day.

See GPU prices and clouds

Source: Rank.ai, Cloud GPU platform rank (recorded AI answers). Data as of Sep 17, 2026.

Collections

Countries, companies, models, GPUs, cities and universities

Each collection ranks one kind of thing on the measures named under it, with the source on every list.

Countries

24 ranked
  1. 1United States0.89
  2. 2China0.80
  3. 3Eurozone0.56

By power, people and prosperity

Companies

500 ranked
  1. 1Amazon$717B
  2. 2Walmart$713B
  3. 3UnitedHealth Group$448B

By revenue, growth and AI visibility

AI models

121 ranked
  1. 1Claude Fable 585
  2. 2GPT-5.6 Sol84
  3. 3Claude Opus 583

By a consensus of 4 leaderboards

GPUs

32 ranked
  1. 1PRO 6000WE 96GB$1.44/hr
  2. 2PRO 6000SE 96GB$2.04/hr
  3. 3A100 80GB$1.49/hr

By price per GPU-hour, by market

US cities

58 ranked
  1. 1Cleveland, OH3.1x
  2. 2Pittsburgh, PA3.2x
  3. 3Oklahoma City, OK3.6x

By jobs, incomes and the cost of a home

Universities

1,148 ranked
  1. 1Massachusetts Institute of Technology$143.4k
  2. 2Massachusetts College of Pharmacy and Health Sciences$125.6k
  3. 3Stanford University$124.1k

By cost, completion and earnings

Your brand

See how often ChatGPT, Claude and Gemini name your brand when asked what to buy.

The AI index

AI models on capability and cost, GPUs on price by market

Two charts from the AI index boards, with every point linked to its board.

Model capability against cost per solved task

DeepSeek V4 Flash solves a task for $0.016. GPT 5.6 Sol Max scores highest, at $0.59 a task.

From the LiveBench release of Jun 25, 2026.

What a GPU-hour costs, by memory and market

An H100 rents for $2.56 an hour on the Akash market. 32 offers across Akash, AWS and Azure.

Akash Oct 16, 2026 · AWS Oct 12, 2026 · Azure Jul 27, 2026

LLM leaderboard

The best AI models on a consensus of Vals AI, Artificial Analysis, LiveBench, LMArena and BFCL, with gateway prices.

Data center map

Planned AI data center capacity by state, with owners, sites and power prices.

Charts

Capex, valuations, compute ownership and citation sources, one chart each.

GPU boards

Every GPU price board, per accelerator, with the evidence behind each row.

See how often AI names you.

Enter your website. In about two minutes, Rank.ai asks ChatGPT, Claude and Gemini 12 questions your customers ask and grades how often they name you.

  • Your grade out of 100How often AI names you, cites your site, and how high it ranks you.
  • Who gets namedEvery competitor in the answers, most named first.
  • The pages AI readsThe sources behind each answer.
  • Three fixesWhat to fix first, with a brief for the first article.