1. Models
  2. Kimi (Moonshot AI)
  3. Vision RL Training Curves on Benchmarks
Kimi (Moonshot AI)X · Jan 30, 2026 · official

Vision RL Training Curves on Benchmarks

Kimi (Moonshot AI)'s small multiples drew 316.8K views on X: 1.4x its median post and #201 of 1,162 AI chart posts we hold by views. Read Oct 13, 2026.

Vision RL training improves visual benchmarks

XDataSmall multiplesKimi agent products and benchmarksAll Kimi (Moonshot AI) models
Published by Kimi (Moonshot AI) on X, Jan 30, 2026. Source: x.com
80/ 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.

#51 of 1,162 AI chart posts we hold · #9 of 20 Kimi (Moonshot AI) 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

Wide but thin: views above the AI chart median, engagement below it, score rank #51.

Its score of 80 ranks #51 of 1,162 AI chart posts, above the median, and #9 of 20 Kimi posts.

Its save rate of 0.21% is in the top decile, 91st percentile, above the 0.076% median.

Its views-per-follower of 0.87 is above the 0.20 median, at the 75th percentile.

12xthe median small multiples post

Across 43 small multiples posts we hold, the median gets 27K views.

2charts in this one post

The numbers belong to the post, so every chart in it shares them and it is ranked once.

top 9%save rate

0.207% of viewers bookmarked it to come back to. The median post gets 0.076%.

2posts from Kimi (Moonshot AI) that day

Part of a run of 2 posts on Jan 30, 2026. This one was #1 of 2 by views that day.

Public X numbers from our read on Oct 13, 2026. Peers: 1,162 AI chart posts we hold and 20 from Kimi (Moonshot AI).

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 #51 of 1,162

Views316.8K#201top 17%
Likes1.9K#168top 14%
Engagement rate0.71%#599bottom 48%
Save rate0.207%#104top 9%

Among 20 Kimi (Moonshot AI) posts

Score #9 of 20

Views316.8K#9top 43%
Likes1.9K#10top 48%
Engagement rate0.71%#13bottom 38%
Save rate0.207%#6top 28%

Growth

Our reads cover 3 days

Views316.8K
Age at our read256 days
Views per day of life1.2Kviews divided by days since it was posted
Gained between our reads+17views over 3 days, about 6 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.

Kimi (Moonshot AI)'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 · Jan 30, 2026

Kimi Vision RL chart scores #51 of 1,162 AI chart posts

A small-multiples chart on Vision RL training curves drew 317K views and a 0.21% save rate.

What happened

Kimi posted 2 charts on January 30, 2026, about Vision RL training curves and token efficiency. The same day it also posted an OSWorld benchmarking chart with 51.5K views.

The numbers

The post got 317K views, ranking #201 of 1,162 AI chart posts, and a 0.21% save rate, ranking #104. Its views-per-follower was 0.87, above the 0.20 median.

The lesson

With 2 charts and a 0.21% save rate, the small-multiples format earned top-decile saves. Use small multiples to boost bookmarking, even when engagement rate lags.

What it means

Its score of 80 places it #9 of 20 Kimi posts, above the account's median. The 0.21% save rate, 91st percentile, shows the format's strength for saving, not liking.

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

Updates: what Kimi (Moonshot AI) posted next

  1. OSWorld Benchmarking Multimodal Agents51.5K views
  2. AI Model Benchmark Comparison7.6M views
  3. LLM Inference on M3 Max, Performance Evolution84K views
  4. Multi-Objective Performance Optimization16K views
  5. Compare Categories by usecase on OpenRouter341.8K views
  6. Weekly Top Models Usage140.2K views

Data and API

100 points read off this chart

Estimated from the image 30%
SeriesGroupValueRead
0.715estimated
0.715estimated
0.715estimated
0.715estimated
0.71estimated
0.71estimated
0.715estimated
0.72estimated
0.72estimated
0.72estimated
0.72estimated
0.72estimated

Showing 12 of 100. All points, the redraw and the maker

Max plan · API

GET https://viral-chart.com/api/posts/a05914e4719e.json
GET https://viral-chart.com/api/charts/a05914e4719e.json
GET https://viral-chart.com/api/charts/a05914e4719e.skill.md
Make one like thisOpen the post JSONChart data JSONSkill fileAPI 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 a05914e4719e.

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 planAPI docs
PathReturnsPlan
GET/api/v1/charts/a05914e4719eTitle, publisher, date, form, point count; with a Max key, the data linkOpen
GET/api/v1/charts/a05914e4719e/dataEvery point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSVMax
GET/api/v1/charts/a05914e4719e/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/a05914e4719e/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 agent products and benchmarks

Kimi agent products and benchmarks: Token Efficiency and Usage Before/After Toggle
OfficialX · Kimi (Moonshot AI)Jan 30, 2026
Token Efficiency and Usage Before/After Toggle316.8K views · 1.9K likes
Data
Kimi agent products and benchmarks: TSLA Stock Price Chart
OfficialX · Kimi (Moonshot AI)Jun 8, 2026
TSLA Stock Price Chart12.4K views · 100 likes
Data
Kimi agent products and benchmarks: AI Model Performance Comparison
OfficialGitHub · Kimi (Moonshot AI)Apr 30, 2026
AI Model Performance Comparison
Data
Kimi agent products and benchmarks: MLLM Performance Comparison Across Categories
OfficialGitHub · Kimi (Moonshot AI)Apr 30, 2026
MLLM Performance Comparison Across Categories
Kimi agent products and benchmarks: Model Confidence vs Proportion
OfficialBlog · Kimi (Moonshot AI)
Model Confidence vs Proportion
Data
Kimi agent products and benchmarks: Execution Time to Solve Tasks of Different Complexity
OfficialBlog · Kimi (Moonshot AI)
Execution Time to Solve Tasks of Different Complexity
Data

More from Kimi (Moonshot AI)

Kimi (Moonshot AI) grouped bar chart: AI model benchmark comparison11x
Kimi (Moonshot AI)·Apr 2026
AI model benchmark comparison
7.6M views17.9K likes0.3% eng.
Data
Kimi (Moonshot AI) grouped bar chart: AI model benchmark comparison9.7x
Kimi (Moonshot AI)·Jan 2026
AI model benchmark comparison
7.4M views15.6K likes0.3% eng.
Data
Kimi (Moonshot AI) grouped bar chart: AI model benchmark comparison8.6x
Kimi (Moonshot AI)·Jun 2026
AI model benchmark comparison
2.7M views13.9K likes0.6% eng.
Data
Kimi K2.7 Code and Kimi Code: AI Model Benchmark Comparison
OfficialX · Kimi (Moonshot AI)Jul 16, 2026
AI Model Benchmark Comparison25.6M views · 56.9K likes
Data
Kimi K2.7 Code and Kimi Code: Coding Benchmarks Comparison
OfficialX · Kimi (Moonshot AI)Jul 16, 2026
Coding Benchmarks Comparison25.6M views · 56.9K likes
Data
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
Data
Kimi K2: Kimi Code Bench Performance vs Tokens
OfficialX · Kimi (Moonshot AI)Jun 12, 2026
Kimi Code Bench Performance vs Tokens2.7M views · 13.9K likes
Data
Kimi K3: AI Model Benchmarks Comparison
OfficialX · Kimi (Moonshot AI)Jul 16, 2026
AI Model Benchmarks Comparison357.6K views · 1.4K likes
Data
Kimi K2: Kimi Code Bench
OfficialX · Kimi (Moonshot AI)Jan 27, 2026
Kimi Code Bench258.9K views · 2.5K likes
Data

Questions people ask

Where does this chart come from?

Kimi (Moonshot AI) published it on Jan 30, 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.