Multimodal · Benchmark profile

CharXiv

Measures reasoning over charts in scientific papers.

Ranking weightedMay 2026Active16 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on CharXiv

Leaderboard · 16 models

Claude Mythos 5 leads at 93.5%, followed by Kimi K3 (91.3%) and Claude Opus 4.8 (89.9%).

Ranking · 4.2%MultimodalMay 2026

#1Closed

Claude Mythos 5

Anthropic

claude-mythos-5

93.5%

Overall #1 · Context 1M tokens

#2Closed

Kimi K3

Moonshot AI

kimi-k3

91.3%

Overall #3 · Context 1M tokens

#3Closed

Claude Opus 4.8

Anthropic

claude-opus-4-8

89.9%

Overall #5 · Context 1M tokens

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

AnthropicMoonshot AIMetaAlibaba CloudGoogleOpenAI

Leaderboard bars

16 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
Claude Mythos 5	#1	Anthropic	93.5%
Kimi K3	#2	Moonshot AI	91.3%
Claude Opus 4.8	#3	Anthropic	89.9%
Muse Spark 1.1	#4	Meta	88.4%
Claude Sonnet 5	#5	Anthropic	88.3%
Muse Spark	#6	Meta	86.4%
Qwen3.7-Plus	#7	Alibaba Cloud	85.9%
Gemini 3.5 Flash	#8	Google	84.2%
GPT-5.5	#9	OpenAI	84.1%
Gemini 3.1 Pro Preview	#10	Google	83.3%
GPT-5.4	#11	OpenAI	82.8%
Qwen3.6 Plus	#12	Alibaba Cloud	81.5%
Gemini 3 Pro	#13	Google	81.4%
Claude Sonnet 4.6	#14	Anthropic	77.4%
Gemini 3.1 Flash-Lite	#15	Google	73.2%
Claude Opus 4.5	#16	Anthropic	68.5%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
Claude Mythos 5	#1	Anthropic	93.5%
Kimi K3	#2	Moonshot AI	91.3%
Claude Opus 4.8	#3	Anthropic	89.9%
Muse Spark 1.1	#4	Meta	88.4%
Claude Sonnet 5	#5	Anthropic	88.3%
Muse Spark	#6	Meta	86.4%
Qwen3.7-Plus	#7	Alibaba Cloud	85.9%
Gemini 3.5 Flash	#8	Google	84.2%
GPT-5.5	#9	OpenAI	84.1%
Gemini 3.1 Pro Preview	#10	Google	83.3%
GPT-5.4	#11	OpenAI	82.8%
Qwen3.6 Plus	#12	Alibaba Cloud	81.5%
Gemini 3 Pro	#13	Google	81.4%
Claude Sonnet 4.6	#14	Anthropic	77.4%
Gemini 3.1 Flash-Lite	#15	Google	73.2%
Claude Opus 4.5	#16	Anthropic	68.5%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Moonshot AI	1	91.3%
Meta	2	87.4%
Alibaba Cloud	2	83.7%
Anthropic	5	83.5%
OpenAI	2	83.4%
Google	4	80.5%

Benchmark score table (16 models)

One best score per model · higher is better

CharXiv leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	Claude Mythos 5 claude-mythos-5	Anthropic	closed	93.5%
#2	Kimi K3 kimi-k3	Moonshot AI	closed	91.3%
#3	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	89.9%
#4	Muse Spark 1.1 muse-spark-1-1	Meta	closed	88.4%
#5	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	88.3%
#6	Muse Spark muse-spark	Meta	closed	86.4%
#7	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	85.9%
#8	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	84.2%
#9	GPT-5.5 gpt-5.5	OpenAI	closed	84.1%
#10	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	83.3%
#11	GPT-5.4 gpt-5.4	OpenAI	closed	82.8%
#12	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	81.5%
#13	Gemini 3 Pro gemini-3-pro	Google	closed	81.4%
#14	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	77.4%
#15	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	73.2%
#16	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	68.5%

The top of this snapshot is led by Claude Mythos 5 at 93.5%; third place is 3.6 points behind. The top-10 spread is 10.2 points.

Compare top models

Claude Mythos 5 vs Kimi K3 Kimi K3 vs Claude Opus 4.8 Claude Opus 4.8 vs Muse Spark 1.1 Muse Spark 1.1 vs Claude Sonnet 5

About CharXiv

Definition and scoring

Organisation: CharXiv
Category: Multimodal
Version: May 2026
Direction: higher is better
Ranking use: Weighted 4.2%
Contamination risk: Unknown

Measures reasoning over charts in scientific papers. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.

Open benchmark source ↗

FAQ

Quick answers

What does CharXiv measure?

Measures reasoning over charts in scientific papers.

Which model scores highest on CharXiv?

Claude Mythos 5 by Anthropic currently leads with 93.5%.

How many models are evaluated on CharXiv?

16 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 4.2% of its capability category. Claude Mythos 5 is overall #1.

MMMU

Rolling · 1 results · Not Weighted

MMMU-Pro

Rolling · 71 results · Weighted 4.2%

MathVista

Rolling · Not evaluated · Not Weighted

ChartQA

Rolling · Not evaluated · Not Weighted

DocVQA

Rolling · Not evaluated · Not Weighted

Blueprint-Bench 2

v2 · 5 results · Reference

Multimodal · Benchmark profile

CharXiv

Measures reasoning over charts in scientific papers.

Ranking weightedMay 2026Active16 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on CharXiv

Leaderboard · 16 models

Claude Mythos 5 leads at 93.5%, followed by Kimi K3 (91.3%) and Claude Opus 4.8 (89.9%).

Ranking · 4.2%MultimodalMay 2026

#1Closed

Claude Mythos 5

Anthropic

claude-mythos-5

93.5%

Overall #1 · Context 1M tokens

#2Closed

Kimi K3

Moonshot AI

kimi-k3

91.3%

Overall #3 · Context 1M tokens

#3Closed

Claude Opus 4.8

Anthropic

claude-opus-4-8

89.9%

Overall #5 · Context 1M tokens

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

AnthropicMoonshot AIMetaAlibaba CloudGoogleOpenAI

Leaderboard bars

16 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
Claude Mythos 5	#1	Anthropic	93.5%
Kimi K3	#2	Moonshot AI	91.3%
Claude Opus 4.8	#3	Anthropic	89.9%
Muse Spark 1.1	#4	Meta	88.4%
Claude Sonnet 5	#5	Anthropic	88.3%
Muse Spark	#6	Meta	86.4%
Qwen3.7-Plus	#7	Alibaba Cloud	85.9%
Gemini 3.5 Flash	#8	Google	84.2%
GPT-5.5	#9	OpenAI	84.1%
Gemini 3.1 Pro Preview	#10	Google	83.3%
GPT-5.4	#11	OpenAI	82.8%
Qwen3.6 Plus	#12	Alibaba Cloud	81.5%
Gemini 3 Pro	#13	Google	81.4%
Claude Sonnet 4.6	#14	Anthropic	77.4%
Gemini 3.1 Flash-Lite	#15	Google	73.2%
Claude Opus 4.5	#16	Anthropic	68.5%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
Claude Mythos 5	#1	Anthropic	93.5%
Kimi K3	#2	Moonshot AI	91.3%
Claude Opus 4.8	#3	Anthropic	89.9%
Muse Spark 1.1	#4	Meta	88.4%
Claude Sonnet 5	#5	Anthropic	88.3%
Muse Spark	#6	Meta	86.4%
Qwen3.7-Plus	#7	Alibaba Cloud	85.9%
Gemini 3.5 Flash	#8	Google	84.2%
GPT-5.5	#9	OpenAI	84.1%
Gemini 3.1 Pro Preview	#10	Google	83.3%
GPT-5.4	#11	OpenAI	82.8%
Qwen3.6 Plus	#12	Alibaba Cloud	81.5%
Gemini 3 Pro	#13	Google	81.4%
Claude Sonnet 4.6	#14	Anthropic	77.4%
Gemini 3.1 Flash-Lite	#15	Google	73.2%
Claude Opus 4.5	#16	Anthropic	68.5%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Moonshot AI	1	91.3%
Meta	2	87.4%
Alibaba Cloud	2	83.7%
Anthropic	5	83.5%
OpenAI	2	83.4%
Google	4	80.5%

Benchmark score table (16 models)

One best score per model · higher is better

CharXiv leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	Claude Mythos 5 claude-mythos-5	Anthropic	closed	93.5%
#2	Kimi K3 kimi-k3	Moonshot AI	closed	91.3%
#3	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	89.9%
#4	Muse Spark 1.1 muse-spark-1-1	Meta	closed	88.4%
#5	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	88.3%
#6	Muse Spark muse-spark	Meta	closed	86.4%
#7	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	85.9%
#8	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	84.2%
#9	GPT-5.5 gpt-5.5	OpenAI	closed	84.1%
#10	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	83.3%
#11	GPT-5.4 gpt-5.4	OpenAI	closed	82.8%
#12	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	81.5%
#13	Gemini 3 Pro gemini-3-pro	Google	closed	81.4%
#14	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	77.4%
#15	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	73.2%
#16	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	68.5%

The top of this snapshot is led by Claude Mythos 5 at 93.5%; third place is 3.6 points behind. The top-10 spread is 10.2 points.

Compare top models

Claude Mythos 5 vs Kimi K3 Kimi K3 vs Claude Opus 4.8 Claude Opus 4.8 vs Muse Spark 1.1 Muse Spark 1.1 vs Claude Sonnet 5

About CharXiv

Definition and scoring

Organisation: CharXiv
Category: Multimodal
Version: May 2026
Direction: higher is better
Ranking use: Weighted 4.2%
Contamination risk: Unknown

Open benchmark source ↗

FAQ

Quick answers

What does CharXiv measure?

Measures reasoning over charts in scientific papers.

Which model scores highest on CharXiv?

Claude Mythos 5 by Anthropic currently leads with 93.5%.

How many models are evaluated on CharXiv?

16 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 4.2% of its capability category. Claude Mythos 5 is overall #1.

MMMU

Rolling · 1 results · Not Weighted

MMMU-Pro

Rolling · 71 results · Weighted 4.2%

MathVista

Rolling · Not evaluated · Not Weighted

ChartQA

Rolling · Not evaluated · Not Weighted

DocVQA

Rolling · Not evaluated · Not Weighted

Blueprint-Bench 2

v2 · 5 results · Reference

Leaderboard · 16 models

Claude Mythos 5

Kimi K3

Claude Opus 4.8

See the field, not just the table.

Benchmark score table (16 models)

Compare top models

Definition and scoring

Quick answers

What does CharXiv measure?

Which model scores highest on CharXiv?

How many models are evaluated on CharXiv?

Does this affect overall Lumina rank?

More Multimodal benchmarks

Leaderboard · 16 models

Claude Mythos 5

Kimi K3

Claude Opus 4.8

See the field, not just the table.

Benchmark score table (16 models)

Compare top models

Definition and scoring

Quick answers

What does CharXiv measure?

Which model scores highest on CharXiv?

How many models are evaluated on CharXiv?

Does this affect overall Lumina rank?

More Multimodal benchmarks