Models
DeepSeek
AI Model Accuracy Comparison Across Benchmarks

DeepSeekX · Jun 17, 2024 · official

AI Model Accuracy Comparison Across Benchmarks

DeepSeek's grouped bar chart drew 489.6K views on X: 1.4x its median post and #150 of 1,162 AI chart posts we hold by views. Read Oct 13, 2026.

Model accuracy comparison across six coding benchmarks

XData Grouped bar chart DeepSeek-Coder All DeepSeek modelsDeepSeek-Coder-V2GPT-4-Turbo-0409Gemini-1.5-ProClaude-3-OpusLlama-3-70BCodestral

DeepSeek: AI Model Accuracy Comparison Across Benchmarks — Published by DeepSeek on X, Jun 17, 2024. Source: x.com

75/ 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.

#116 of 1,162 AI chart posts we hold · #6 of 18 DeepSeek 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 at #116 of 1,162 AI chart posts.

Its score of 75 lands at #116 of 1,162 AI chart posts and #6 of 18 DeepSeek posts.

A 0.15% save rate sits at the 79th percentile among AI chart posts, 1.9x their 0.076% median.

At 0.43 views per follower it runs 2.1x the 0.20 median across posts with known followers.

5.6xthe median grouped bar chart post

Across 108 grouped bar chart posts we hold, the median gets 87.5K views.

top 21%save rate

0.147% of viewers bookmarked it to come back to. The median post gets 0.076%.

bottom 21%engagement rate

0.43% of viewers liked, reposted, quoted or replied. The median post gets 0.73%.

0.43views per follower

489.6K views from an account with 1.1M followers.

Public X numbers from our read on Oct 13, 2026. Peers: 1,162 AI chart posts we hold and 18 from DeepSeek.

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 #116 of 1,162

Views	489.6K	#150	top 13%
Likes	1.6K	#192	top 16%
Engagement rate	0.43%	#921	bottom 21%
Save rate	0.147%	#241	top 21%

Among 18 DeepSeek posts

Score #6 of 18

Views	489.6K	#8	top 42%
Likes	1.6K	#8	top 42%
Engagement rate	0.43%	#7	top 36%
Save rate	0.147%	#3	top 14%

Growth

Our reads cover 22 hours

Views489.6K

Age at our read848 days

Views per day of life577views divided by days since it was posted

Gained between our reads+6views over 22 hours, about 7 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.

DeepSeek'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 · Jun 17, 2024

DeepSeek posts six-model coding benchmark chart

A grouped bar chart of accuracy across six coding benchmarks drew 490K views and a 0.43% engagement rate.

What happened

DeepSeek published a grouped bar chart on June 17, 2024 comparing model accuracy across six coding benchmarks. It was the account's only post that day.

The numbers

It reached 490K views, #150 of 1,162 AI chart posts, against a 39.9K median among AI chart posts. Its score of 75 ranks #116 of 1,162 AI chart posts and #6 of 18 DeepSeek posts.

The lesson

The chart shows HumanEval accuracy from 90.2% to 81.1% across six models, and its 0.15% save rate is 1.9x the 0.076% median among AI chart posts. A grouped bar chart with printed values is what readers bookmark at that rate.

What it means

At #6 of 18 DeepSeek posts, this chart sits below the account's own top posts while its 0.43% engagement rate trails the AI chart median. The 490K views against a 39.9K median show the gap between reach and engagement for this publisher.

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

Updates: what DeepSeek posted next

Data and API

42 points read off this chart

Estimated from the image 79%

Series	Group	Value	Read
DeepSeek-Coder-V2	HumanEval	90.2 %	printed
GPT-4-Turbo-0409	HumanEval	89.2 %	printed
Gemini-1.5-Pro	HumanEval	84.9 %	printed
Claude-3-Opus	HumanEval	83.5 %	printed
Llama-3-70B	HumanEval	81.7 %	printed
Codestral	HumanEval	81.1 %	printed
DeepSeek-Coder-V2	MBPP+	76.2 %	printed
GPT-4-Turbo-0409	MBPP+	72.2 %	printed
Gemini-1.5-Pro	MBPP+	74.6 %	printed
Claude-3-Opus	MBPP+	72 %	printed
Llama-3-70B	MBPP+	69 %	printed
Codestral	MBPP+	68.2 %	printed

Showing 12 of 42. All points, the redraw and the maker

Max plan · API

GET https://viral-chart.com/api/posts/e85a6a6fd6a0.json
GET https://viral-chart.com/api/charts/e85a6a6fd6a0.json
GET https://viral-chart.com/api/charts/e85a6a6fd6a0.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 e85a6a6fd6a0.

See the Max plan API docs

	Path	Returns	Plan
GET	`/api/v1/charts/e85a6a6fd6a0`	Title, publisher, date, form, point count; with a Max key, the data link	Open
GET	`/api/v1/charts/e85a6a6fd6a0/data`	Every point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSV	Max
GET	`/api/v1/charts/e85a6a6fd6a0/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/e85a6a6fd6a0/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 DeepSeek-Coder

DeepSeek-Coder: AI Model MATH Accuracy Over Time

OfficialX · DeepSeekFeb 6, 2024

AI Model MATH Accuracy Over Time257.1K views · 1K likes