AI model task length
METR's scatter plot drew 270.5K views on X: 1.8x its typical post and #8 of 26 METR charts we track by views. Read Oct 10, 2026.
We ran GPT-5.4 (xhigh) on our tasks. Its time-horizon depends greatly on our treatment of reward hacks: the point estimate would be 5.7hrs (95% CI of 3hrs to 13.5hrs) under our standard methodology, but 13hrs (95% CI of 5hrs to 74hrs) if we
Why it spread
It spread because the post explains itself and drew many quotes, with 771 likes, 59 reposts, 39 quotes, 177 saves and 270,452 views on a scatter chart about AI model task length growing exponentially over time.
What worked
- The post explains itself, so readers could grasp the reward hack comparison without extra context.
- It drew many quotes, which pushed conversation and further sharing beyond the original audience.
- The chart is self labelling and branded with a source line, which made it easy to trust and reuse.
What to borrow
We could pair one of our own charts with a plain sentence that states the headline result and the range around it, then invite readers to quote it with their own interpretation.
Careful
Do not over-read the quote rate or view count as proof of durable impact, and do not copy the reward hack framing without checking whether our own data supports it.
Written by our analysis model from the measured numbers above, not by METR.
Make it yours
Your numbers, the bar chart way.
Starts from the real numbers of "Task-5.4 (xhigh) has a 50% success-time-horizon of about 5.7 hours (95% CI: 3 hrs to 13.5 " by METR. 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 63c1ded30985.
Approximate values. Our two readings of this chart agreed only in part (confidence 0.39 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/63c1ded30985 | Title, publisher, date, form, point count; with a Max key, the data link | Open |
| GET | /api/v1/charts/63c1ded30985/data | Every point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSV | Max |
| GET | /api/v1/charts/63c1ded30985/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/63c1ded30985/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 METR
All 26
Other giants using a scatter plot
All 78
Questions people ask
How far did METR's "AI model task length" chart travel?
270,452 views, 771 likes, 59 reposts and 177 saves on X, read Oct 10, 2026. A normal post from @METR_Evals gets about 153K views.
What kind of chart is it?
A scatter plot. AI model task length grows exponentially over time.
Can I use METR's chart?
Use the format, not the picture. The chart, logo and design belong to METR. ViralChart redraws the chart type with your numbers and words, and credits the original publisher.