Charts
Output Reasoning vs Answer Tokens for AI Models

@ArtificialAnlysX · Mar 20, 2026

Output Reasoning vs Answer Tokens for AI Models

@ArtificialAnlys's bar chart drew 4.1K views on X: 0.4x @artificialanlys's median post and #985 of 1,162 AI chart posts we hold by views. Read Oct 12, 2026.

Output reasoning and answer tokens across AI models

XData Bar chart Mistral SmallQwen 3 598 BNVIDIA Nemotron Nano 2NVIDIA Nemotron SuperQwen 3 235 B A22BQwen 3 120B A22BClaude 4.5 Sonnetgpt-oss-120b (HF)GLM-4.5-Flash

@ArtificialAnlys: Output Reasoning vs Answer Tokens for AI Models — Published by @ArtificialAnlys on X, Mar 20, 2026. Source: x.com

13/ 100

ViralChart score

Quiet

Fewer views and a lower engagement rate than the median post.

#1,135 of 1,162 AI chart posts we hold · #474 of 491 @artificialanlys 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 Quiet quadrant, at #1,135 of 1,162 AI chart posts.

Its save rate of 0.000% sits at the 2nd percentile among AI chart posts, in the bottom decile.

A score of 13 places it at #1,135 of 1,162 AI chart posts.

At 4.1K views it runs 0.36x this company's median of 11.4K views.

bottom 2%save rate

0.000% of viewers bookmarked it to come back to. The median post gets 0.076%.

0.2xthe median bar chart post

Across 241 bar chart posts we hold, the median gets 25.6K views.

0.4x@artificialanlys's median post

4.1K views, where @artificialanlys's typical post gets 11.4K.

3posts from @artificialanlys that day

Part of a run of 3 posts on Mar 20, 2026. This one was #3 of 3 by views that day.

Public X numbers from our read on Oct 12, 2026. Peers: 1,162 AI chart posts we hold and 491 from @artificialanlys.

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 #1,135 of 1,162

Views	4.1K	#985	bottom 15%
Likes	18	#1,041	bottom 10%
Engagement rate	0.48%	#861	bottom 26%
Save rate	0.000%	#1,128	bottom 2%

Among 491 @artificialanlys posts

Score #474 of 491

Views	4.1K	#361	bottom 27%
Likes	18	#401	bottom 18%
Engagement rate	0.48%	#390	bottom 21%
Save rate	0.000%	#464	bottom 3%

Growth

One read so far

Views4.1K

Age at our read206 days

Views per day of life20views 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.

@artificialanlys'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 · Mar 20, 2026

@ArtificialAnlys charts output reasoning and answer tokens

A 13-model bar chart on output reasoning and answer tokens drew 4.1K views and a 0.000% save rate.

What happened

On March 20, 2026, @ArtificialAnlys posted a bar chart on output reasoning and answer tokens across AI models. The account posted 3 items that day, including Artificial Analysis Intelligence Index v4.0 and Intelligence vs. Total Parameters.

The numbers

The post reached 4.1K views, ranking #985 of 1,162 AI chart posts, against a median of 39.9K views among AI chart posts. Its score of 13 ranks #1,135 of 1,162 AI chart posts and #474 of 491 @ArtificialAnlys posts.

The lesson

A 0.000% save rate at the 2nd percentile among AI chart posts: this 13-model bar chart gave readers nothing to bookmark. The chart shows Reasoning Tokens (NVIDIA Nemotron 3 Nano) at 140.

What it means

At 0.36x this company's median views, the post ranks #361 of 491 @ArtificialAnlys posts by views. The gap to the 39.9K median among AI chart posts is 4.1K against 39.9K.

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

Updates: what @artificialanlys posted next

Data and API

33 points read off this chart

Estimated from the image 39%

Series	Group	Value	Read
Reasoning Tokens	Qwen3.5-58B Q4	190 M	estimated
Reasoning Tokens	NVIDIA Nemotron 3 Nano	140 M	printed
Reasoning Tokens	NVIDIA Nemotron 3 Super	110 M	estimated
Reasoning Tokens	Qwen3.5-235B A28B	100 M	estimated
Reasoning Tokens	Qwen3.5-235B A70B	90 M	estimated
Reasoning Tokens	Claude 4.5 Haiku	82 M	estimated
Reasoning Tokens	gpt-oss-120B (high)	79 M	estimated
Reasoning Tokens	GLM-4.7-Flash	78 M	estimated
Reasoning Tokens	gpt-oss-20B (high)	64 M	estimated
Reasoning Tokens	Mistral Small 4	61 M	estimated
Answer Tokens	Claude 4.5 Haiku	56 M	estimated
Answer Tokens	Mistral Large 3	52 M	estimated

Showing 12 of 33. All points, the redraw and the maker

Max plan · API

GET https://viral-chart.com/api/posts/ae77fadbda6a.json
GET https://viral-chart.com/api/charts/ae77fadbda6a.json
GET https://viral-chart.com/api/charts/ae77fadbda6a.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 ae77fadbda6a.

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.

See the Max plan API docs

	Path	Returns	Plan
GET	`/api/v1/charts/ae77fadbda6a`	Title, publisher, date, form, point count; with a Max key, the data link	Open
GET	`/api/v1/charts/ae77fadbda6a/data`	Every point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSV	Max
GET	`/api/v1/charts/ae77fadbda6a/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/ae77fadbda6a/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 Mistral Small

OfficialHugging Face · Mistral AIJul 15, 2026

Score vs. Output Length

Data

OfficialHugging Face · Mistral AIJul 15, 2026

Performance Comparison Across Internal Models

Data

OfficialHugging Face · Mistral AIJul 15, 2026

Score vs. Output Length

Data

OfficialHugging Face · Mistral AIJul 15, 2026

Score vs. Output Length

Data

OfficialBlog · Mistral AIMar 17, 2025

Performance / Mistral GPQA-Diamond

OfficialBlog · Mistral AIJan 30, 2025

Performance / MMLU-Pro vs Latency

More from Mistral AI

OfficialX · Mistral AIDec 2, 2025

LMArena ELO Score441.6K views · 572 likes

Data

Mistral Large: Artificial Analysis Cyber Index v1

OfficialX · Mistral AIOct 6, 2026

Artificial Analysis Cyber Index v1134.5K views · 2.8K likes

Data

Mistral Large: Artificial Analysis - AutomationBench

OfficialX · Mistral AIOct 7, 2026

Artificial Analysis - AutomationBench42.7K views · 863 likes

Data

Mistral Large: Vals.ai - Harvey's Legal Agent Benchmark

OfficialX · Mistral AIOct 7, 2026

Vals.ai - Harvey's Legal Agent Benchmark7.1K views · 111 likes

Data

OfficialBlog · Mistral AIOct 6, 2026

Artificial Analysis - DeepSWE 1.1

OfficialBlog · Mistral AIOct 6, 2026

Code Benchmarks - Terminal-Bench 4.0

Questions people ask

Where does this chart come from?

@ArtificialAnlys published it on Mar 20, 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.