- Giants
- SemiAnalysis
- Token throughput over time
Token throughput over time
SemiAnalysis's line chart drew 198.2K views on X: 4.9x @semianalysis_'s median post and #267 of 1,162 AI chart posts we hold by views. Read Oct 10, 2026.
Speed is the Moat! Measuring performance over time is what matters. For single-node FP4 MI355 DeepSeek R1 MoE, every time we update the SGLang version, we continue to see rapid increases in performance! Check out the latest perf at inferenc
Why it spread
It spread because an independent author with a big source posted a line chart showing token throughput per GPU rising over time, and it reached 198,180 views with 209 likes, 11 reposts, 6 quotes and 61 saves.
What worked
- The post frames the chart around a simple competitive idea, that speed is the moat and that measuring performance over time is what matters.
- The chart is a line showing token throughput per GPU increasing over time, which makes steady progress easy to read at a glance.
- The author is independent and the source reach sits in the top quarter, while the quote rate is above average, so the claim traveled through discussion rather than only passive viewing.
What to borrow
We could take our own repeated benchmark runs across versions and plot one clear line of throughput per unit of hardware over time, then lead with a plain framing sentence about why that trend matters.
Careful
Do not copy the low design polish or the missing source line, and do not over-read the engagement: the saves and quotes are solid but the reposts are modest, so this is not proof that the chart alone drove the spread.
Written by our analysis model from the measured numbers above, not by SemiAnalysis.
Make it yours
Your numbers, in this line chart.
Starts from the real numbers of "Interpolated Token Throughput per GPU (tok/s/gpu)" by semianalysis. 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 dad804718245.
| Path | Returns | Plan | |
|---|---|---|---|
| GET | /api/v1/charts/dad804718245 | Title, publisher, date, form, point count; with a Max key, the data link | Open |
| GET | /api/v1/charts/dad804718245/data | Every point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSV | Max |
| GET | /api/v1/charts/dad804718245/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/dad804718245/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 SemiAnalysis
All 33
Other giants using a line chart
All 295
Questions people ask
How far did SemiAnalysis's "Token throughput over time" chart travel?
198,180 views, 209 likes, 11 reposts and 61 saves on X, read Oct 10, 2026. A normal post from @SemiAnalysis_ gets about 52.2K views.
What kind of chart is it?
A line chart. Token throughput per GPU increases over time.
Can I use SemiAnalysis's chart?
Use the format, not the picture. The chart, logo and design belong to SemiAnalysis. ViralChart redraws the chart type with your numbers and words, and credits the original publisher.