1. Models
  2. xAI
  3. Agent Arena Leaderboard
a
@arenaX · Sep 25, 2026

Agent Arena Leaderboard

@arena's benchmark bar chart drew 123.6K views on X: 2.7x its median post and #343 of 1,162 AI chart posts we hold by views. Read Oct 13, 2026.

Grok-4.7 shows 3.9% net improvement in Agent Arena

XDataBenchmark bar chartGrok 4All xAI modelsClaude Fable 5.1 (Max)GPT-6 Astra (Max)Claude Opus 5 (Max)Claude Fable 5 (High)Claude Opus 5 (High)Claude Opus 4.8 (High)GPT-5.6 Sol (xHigh)Claude Sonnet 5 (High)
Published by @arena on X, Sep 25, 2026. Source: x.com
54/ 100

ViralChart score

Wide but thin

More views than the median post but a lower engagement rate: many saw it, few acted. Common for big accounts.

#555 of 1,162 AI chart posts we hold · #124 of 239 @arena posts. Score = views 40%, likes 30%, engagement rate 15%, save rate 15%, each as a percentile of every AI chart post we hold.

Our read

The post sits in the Wide but thin quadrant, with a score rank of #555 of 1,162 AI chart posts.

Its 124K views are 2.7x @arena's median post, ranking #65 of 239 @arena posts by views.

Saves landed at 0.041%, the 22nd percentile among AI chart posts and 0.54x their 0.076% median.

It was one of 4 posts from this account that day, ranking #3 of 4 by views.

4posts from @arena that day

Part of a run of 4 posts on Sep 25, 2026. This one was #3 of 4 by views that day.

2.7x@arena's median post

123.6K views, where @arena's typical post gets 46.2K.

bottom 22%save rate

0.041% of viewers bookmarked it to come back to. The median post gets 0.076%.

bottom 25%engagement rate

0.47% of viewers liked, reposted, quoted or replied. The median post gets 0.73%.

Public X numbers from our read on Oct 13, 2026. Peers: 1,162 AI chart posts we hold and 239 from @arena.

Where it sits

Reach against resonance

Each dot is a post. Right is more views, up is a higher engagement rate. The lines are the medians, so the four corners are the four kinds of post.

Among 1,162 AI chart posts we hold

Score #555 of 1,162

Views123.6K#343top 29%
Likes495#445top 38%
Engagement rate0.47%#873bottom 25%
Save rate0.041%#905bottom 22%

Among 239 @arena posts

Score #124 of 239

Views123.6K#65top 27%
Likes495#87top 36%
Engagement rate0.47%#178bottom 26%
Save rate0.041%#199bottom 17%

Growth

Our reads cover 27 hours

Views123.6K
Age at our read19 days
Views per day of life6.6Kviews divided by days since it was posted
Gained between our reads+28views over 27 hours, about 25 a day now

Dots are our reads. The dotted line from the day it was posted to our first read is a gap, not data.

@arena's posts over time

Every post from this account we hold, by date and views. This one is marked; the line is the running median of the last 10.

News · Sep 25, 2026

@arena posts Agent Arena Leaderboard chart on September 25, 2026

A 20-model benchmark bar chart drew 124K views but a 0.041% save rate, below the AI chart median.

What happened

@arena published a Benchmark bar chart on the Agent Arena Leaderboard on September 25, 2026. It was one of 4 posts from this account that day, alongside How Claude Models Write Language Complexity, Agent Arena Leaderboard Rankings and GPT-6Sol (Max) Pareto Frontier for Agent Arena.

The numbers

It reached 124K views, #343 of 1,162 AI chart posts, against a median of 39.9K. Its score of 54 ranked #555 of 1,162 AI chart posts and #124 of 239 @arena posts.

The lesson

The chart shows 20 models, and saves came in at 0.041%, 0.54x the 0.076% median among AI chart posts. A 20-model comparison drew 124K views but a 22nd percentile save rate.

What it means

At #124 of 239 @arena posts by score, this chart post sits below the account's own top tier despite 2.7x its median views. The 0.041% save rate, 0.54x the AI chart median, marks it as wide but thin.

Written by ViralChart from public numbers read on Oct 13, 2026. Every number is checked against our data before it is shown.

Data and API

19 points read off this chart

Estimated from the image 83%
SeriesGroupValueRead
Claude Fable 5.1 (Max)13.4 %printed
GPT-6 Astra (Max)11 %printed
Claude Opus 5 (Max)9.5 %printed
Claude Fable 5 (High)8.3 %printed
Claude Opus 4.8 (High)6.2 %printed
GPT-5.6 Sol (xHigh)4.8 %printed
Claude Sonnet 5 (High)4.5 %printed
Kimi K3 (Max)4.4 %printed
GPT 5.5 (xHigh)4.3 %printed
Hy4 Preview4 %printed
DeepSeek-V4.1 Flash Max4 %printed
Muse Spark 1.3 (Max)4 %printed

Showing 12 of 19. All points, the redraw and the maker

Max plan · API

GET https://viral-chart.com/api/posts/8d0185389f9d.json
GET https://viral-chart.com/api/charts/8d0185389f9d.json
GET https://viral-chart.com/api/charts/8d0185389f9d.skill.md

Static files in this prototype. The post JSON holds the public numbers, ranks, score, quadrant, our reads and, when checked, our copy.

API · Max

Use this chart via API

On the Max plan you can pull this chart's numbers with the provenance of each one, as JSON or CSV, with its redraw spec, or call it from an agent over MCP. Chart id 8d0185389f9d.

PathReturnsPlan
GET/api/v1/charts/8d0185389f9dTitle, publisher, date, form, point count; with a Max key, the data linkOpen
GET/api/v1/charts/8d0185389f9d/dataEvery point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSVMax
GET/api/v1/charts/8d0185389f9d/render?view=1The draw-only view this page usesOpen
POST/mcpTool get_chart_data with this id, for Claude, Cursor and other agentsMax
curlMax
curl -s https://viral-chart.com/api/v1/charts/8d0185389f9d/data \
  -H "Authorization: Bearer $VIRALCHART_KEY"
Claude CodeMax
claude mcp add --transport http viralchart https://viral-chart.com/mcp \
  --header "Authorization: Bearer $VIRALCHART_KEY"

60 requests a minute per key; over that you get 429 with Retry-After. Up to 5 keys per account, made on your account page and shown once. Without Max, the data route answers 402 with a link to the plans. The numbers describe a chart its publisher owns, and every file carries a license note.

More charts about Grok 4

More from xAI

Questions people ask

Where does this chart come from?

@arena published it on Sep 25, 2026 on X (x.com). We keep it here, with the publisher and date credited as text, so you can study it without leaving.

Can I make a chart like this?

Use the format, not the picture. Open its data page and use Make one like this: it starts from this chart's numbers and takes yours.