Models
Kimi (Moonshot AI)
Token Throughput per GPU vs. Interactivity

@SemiAnalysis_X · Apr 16, 2026

Token Throughput per GPU vs. Interactivity

@SemiAnalysis_'s line chart drew 52.1K views on X: 1.3x @semianalysis_'s median post and #523 of 1,162 AI chart posts we hold by views. Read Oct 12, 2026.

Token throughput vs interactivity for GB200 NVL72

XData Line chart Kimi K2 All Kimi (Moonshot AI) modelsGB200 NVL72 (Dynamo vLLM)

@SemiAnalysis_: Token Throughput per GPU vs. Interactivity — Published by @SemiAnalysis_ on X, Apr 16, 2026. Source: x.com

45/ 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.

#688 of 1,162 AI chart posts we hold · #16 of 24 @semianalysis_ 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 #688 of 1,162 AI chart posts.

Its 0.38% engagement rate sits at the 16th percentile among AI chart posts, below the 0.73% median.

The score of 45 ranks #688 of 1,162 AI chart posts and #16 of 24 @SemiAnalysis_ posts.

At 52.1K views it runs 1.3x @SemiAnalysis_'s 40.3K company median, ranking #9 of 24 @SemiAnalysis_ posts.

bottom 16%engagement rate

0.38% of viewers liked, reposted, quoted or replied. The median post gets 0.73%.

1.3x@semianalysis_'s median post

52.1K views, where @semianalysis_'s typical post gets 40.3K.

0.7xthe median line chart post

Across 68 line chart posts we hold, the median gets 73.7K views.

top 44%save rate

0.088% of viewers bookmarked it to come back to. The median post gets 0.076%.

Public X numbers from our read on Oct 12, 2026. Peers: 1,162 AI chart posts we hold and 24 from @semianalysis_.

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 #688 of 1,162

Views	52.1K	#523	top 45%
Likes	164	#690	bottom 41%
Engagement rate	0.38%	#978	bottom 16%
Save rate	0.088%	#510	top 44%

Among 24 @semianalysis_ posts

Score #16 of 24

Views	52.1K	#9	top 35%
Likes	164	#15	bottom 40%
Engagement rate	0.38%	#20	bottom 19%
Save rate	0.088%	#16	bottom 35%

Growth

One read so far

Views52.1K

Age at our read179 days

Views per day of life291views divided by days since it was posted

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

@semianalysis_'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 · Apr 16, 2026

@SemiAnalysis_ charts GB200 NVL72 token throughput vs interactivity

A line chart on token throughput per GPU against interactivity drew 52.1K views and a 0.38% engagement rate, below the 0.73% AI chart median.

What happened

@SemiAnalysis_ posted a line chart on token throughput per GPU vs. interactivity for GB200 NVL72 on April 16, 2026. It was the account's only post that day.

The numbers

It reached 52.1K views, #523 of 1,162 AI chart posts by views and #9 of 24 @SemiAnalysis_ posts. Its score of 45 ranks #688 of 1,162 AI chart posts.

The lesson

At 0.38% engagement against a 0.73% AI chart median, this line chart drew 52.1K views but a 16th percentile engagement rate. A single-model line chart can travel on views while sitting below the median on engagement.

What it means

At #16 of 24 @SemiAnalysis_ posts by score, this chart trails the account's own median engagement of 0.62%. Its 1.3x lead over the 40.3K company median views comes with a 0.38% engagement rate, below the 0.73% AI chart median.

Written by ViralChart from public numbers read on Oct 12, 2026. Every number is checked against our data before it is shown.

Updates: what @semianalysis_ posted next

Data and API

18 points read off this chart

Estimated from the image 30%

Series	Group	Value	Read
GB200 (vLLM)	5xDEP4+1xDEP8	12.5 tok/s/gpu	estimated
GB200 (vLLM)	3xDEP4+1xDEP16	7.5 tok/s/gpu	estimated
GB200 (vLLM)	3xDEP4+1xDEP16	6.5 tok/s/gpu	estimated
GB200 (vLLM)	1xDEP4+1xDEP16	4.5 tok/s/gpu	estimated
GB200 (vLLM)	1xDEP4+1xDEP16	3.5 tok/s/gpu	estimated
GB200 (vLLM)	1xDEP4+1xDEP16	2.5 tok/s/gpu	estimated
GB200 (vLLM)	1xDEP4+1xDEP16	1.8 tok/s/gpu	estimated
GB200 (vLLM)	1xDEP4+1xDEP16	1.2 tok/s/gpu	estimated
GB200 (vLLM)	1xDEP4+1xDEP16	0.8 tok/s/gpu	estimated
GB200 (vLLM)	1xDEP4+1xDEP16	0.5 tok/s/gpu	estimated
GB200 (vLLM)	1xDEP4+1xDEP16	0.3 tok/s/gpu	estimated
B200 (vLLM)	TEP4	4 tok/s/gpu	estimated

Showing 12 of 18. All points, the redraw and the maker

Max plan · API

GET https://viral-chart.com/api/posts/96a479fc564a.json
GET https://viral-chart.com/api/charts/96a479fc564a.json
GET https://viral-chart.com/api/charts/96a479fc564a.skill.md

Make one like this Open the post JSON Chart data JSON Skill file API docs

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 96a479fc564a.

Approximate values. Our two readings of this chart agreed only in part (confidence 0.30 of 1), so check the original before you quote a number. The API marks it lowConfidence, and ?min_confidence=0.5 leaves it out.

See the Max plan API docs

	Path	Returns	Plan
GET	`/api/v1/charts/96a479fc564a`	Title, publisher, date, form, point count; with a Max key, the data link	Open
GET	`/api/v1/charts/96a479fc564a/data`	Every point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSV	Max
GET	`/api/v1/charts/96a479fc564a/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

curlMax

curl -s https://viral-chart.com/api/v1/charts/96a479fc564a/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 Kimi K2

Kimi K2: Kimi K2 Thinking vs GPT-5 vs Claude Sonnet

OfficialX · Kimi (Moonshot AI)Nov 6, 2025

Kimi K2 Thinking vs GPT-5 vs Claude Sonnet4.9M views · 9.5K likes