AI model replication performance
Epoch AI's grouped bar chart drew 13.1K views on X: 2.5x @epochairesearch's median post and #779 of 1,162 AI chart posts we hold by views. Read Oct 14, 2026.
Unfortunately, newer models have seen the original innovation during their training, making the task significantly easier. However, even with this information, GPT-6 Astra and Claude Fable 5.1 struggled to reimplement it.
Why it spread
The post spread because it explained itself and rode a hot topic that week, reaching 13,118 views with 124 likes, 2 reposts, 3 quotes, and 12 saves, while the account's typical likes are 114.
What worked
- The post text explains the chart's point directly: newer models saw the original innovation during training, yet GPT-6 Astra and Claude Fable 5.1 still struggled to reimplement it.
- The chart is a grouped bar with high self labelling and a source line, so viewers can understand the comparison without extra context.
- The topic was hot that week and the account is branded, which helped the post travel beyond its usual audience.
What to borrow
We could use the same self explaining grouped bar and plain text framing to show where our own models fail even when the answer was likely in training, then label the source clearly.
Careful
Do not over read the low quote rate or the modest reposts and saves, and do not copy the assumption that a hot topic alone will carry a chart.
Written by our analysis model from the measured numbers above, not by Epoch AI.
Make it yours
Your numbers, in this benchmark bar chart.
Starts from the real numbers of "Even with information from the paper, models struggled to fully replicate its results" by Epoch AI. Replace every number with yours before you post. Make one like this, step by step
Free pictures carry a small "Made with ViralChart" line. That line is how other people find us.
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 df1c0788a36b.
Approximate values. Our two readings of this chart agreed only in part (confidence 0.45 of 1), so check the original before you quote a number. The API marks it lowConfidence, and ?min_confidence=0.5 leaves it out.
| Path | Returns | Plan | |
|---|---|---|---|
| GET | /api/v1/charts/df1c0788a36b | Title, publisher, date, form, point count; with a Max key, the data link | Open |
| GET | /api/v1/charts/df1c0788a36b/data | Every point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSV | Max |
| GET | /api/v1/charts/df1c0788a36b/render?view=1 | The draw-only view this page uses | Open |
| POST | /mcp | Tool get_chart_data with this id, for Claude, Cursor and other agents | Max |
curl -s https://viral-chart.com/api/v1/charts/df1c0788a36b/data \
-H "Authorization: Bearer $VIRALCHART_KEY"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 from Epoch AI
All 97
Other giants using a grouped bar chart
All 50
Questions people ask
How far did Epoch AI's "AI model replication performance" chart travel?
13,199 views, 129 likes, 2 reposts and 12 saves on X, read Oct 14, 2026. A normal post from @EpochAIResearch gets about 68.2K views.
What kind of chart is it?
A grouped bar chart. Models struggle to replicate paper results even with information.
Can I use Epoch AI's chart?
Use the format, not the picture. The chart, logo and design belong to Epoch AI. ViralChart redraws the chart type with your numbers and words, and credits the original publisher.