ERQA

Accessible data for ERQA leaderboard
Model	Rank	Model ID	Provider	Score	Unit
Qwen3.8-Flash-Next	#1	qwen-3-8-flash-next	Alibaba Cloud	72.3	percent
Qwen3.7-Plus	#2	qwen-3-7-plus	Alibaba Cloud	69.8	percent
Qwen3.8-27B	#3	qwen-3-8-27b	Alibaba Cloud	65.5	percent
GPT-5.4	#4	gpt-5-4	OpenAI	65.4	percent
Muse Spark	#5	muse-spark	Meta	64.7	percent
Qwen3.6 27B	#6	qwen3-6-27b	Alibaba Cloud	62.5	percent
Grok 4.20	#7	grok-4-20	xAI	54.1	percent
Claude Opus 4.6	#8	claude-opus-4-6	Anthropic	51.6	percent

ERQA leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Qwen3.8-Flash-Next Qwen/Qwen3.8-Flash-Next	Alibaba Cloud	open	Public reference	72.3%
#2	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	closed	Source carrier	69.8%
#3	Qwen3.8-27B Qwen/Qwen3.8-27B	Alibaba Cloud	open	Public reference	65.5%
#4	GPT-5.4 gpt-5.4	OpenAI	closed	Source carrier	65.4%
#5	Muse Spark muse-spark	Meta	closed	Source carrier	64.7%
#6	Qwen3.6 27B qwen3.6-27b	Alibaba Cloud	open	Source carrier	62.5%
#7	Grok 4.20 grok-4.20	xAI	closed	Source carrier	54.1%
#8	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	Source carrier	51.6%

Rank

Model

Provider

License

Evidence use

Score

Qwen3.8-Flash-Next

Qwen/Qwen3.8-Flash-Next

Alibaba Cloud

open

Public reference

72.3%

Qwen3.7-Plus

qwen3.7-plus

Alibaba Cloud

closed

Source carrier

69.8%

Qwen3.8-27B

Qwen/Qwen3.8-27B

Alibaba Cloud

open

Public reference

65.5%

GPT-5.4

gpt-5.4

OpenAI

closed

Source carrier

65.4%

Muse Spark

muse-spark

Every distinct dated source-published row for ERQA, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Qwen3.8-Flash-Next (xhigh default thinking configuration) qwen-3-8-flash-next-xhigh	Qwen3.8-Flash-Next	72.3%	2026system:qwen3-8-flash-next:erqa	Public reference · not scored source-checked	Qwen3.8-Flash-Next current official launch page ↗Observed 2026-08-26 · checked 2026-08-29
Qwen3.7-Plus (provider-published configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table qwen-3-7-plus-unspecified	Qwen3.7-Plus	69.8%	2026system:qwen3-8-flash-next-comparison:cell:vision:erqa:2	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
Qwen3.8-27B (xhigh default thinking configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table qwen-3-8-27b-xhigh	Qwen3.8-27B	65.5%	2026system:qwen3-8-flash-next-comparison:cell:vision:erqa:1	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
Claude Opus 4.6 (Max) as published in Qwen's Qwen3.8-Flash-Next comparison table claude-opus-4-6-max	Claude Opus 4.6	40.8%	2026system:qwen3-8-flash-next-comparison:cell:vision:erqa:3	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
Qwen3.7-Plus as published by Qwen qwen-3-7-plus-unspecified	Qwen3.7-Plus	69.8%	2026Qwen3.8-27B official VL table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Qwen3.8-27B (xhigh) qwen-3-8-27b-xhigh	Qwen3.8-27B	65.5%	2026Qwen3.8-27B official VL table	Public reference · not scored provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Qwen3.6-27B as published by Qwen qwen3-6-27b-default	Qwen3.6 27B	62.5%	2026Qwen3.8-27B official VL table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Claude Opus 4.6 Max as published by Qwen claude-opus-4-6-max	Claude Opus 4.6	40.8%	2026Qwen3.8-27B official VL table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	69.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	65.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	64.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	62.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	54.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	51.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	69.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	65.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	64.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	62.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	54.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	51.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	69.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	65.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	64.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	62.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	54.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	51.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

What does ERQA measure?

A grounded visual reasoning benchmark focused on evidence-based question answering over real images.

Which model scores highest on ERQA?

Qwen3.8-Flash-Next by Alibaba Cloud currently leads with 72.3%.

How many models are evaluated on ERQA?

8 canonical model products have a qualifying score, backed by 26 distinct dated source rows.

Does this affect overall Lumina rank?

No. This benchmark is display-only and does not enter the overall Lumina composite.

Accessible data for ERQA leaderboard
Model	Rank	Model ID	Provider	Score	Unit
Qwen3.8-Flash-Next	#1	qwen-3-8-flash-next	Alibaba Cloud	72.3	percent
Qwen3.7-Plus	#2	qwen-3-7-plus	Alibaba Cloud	69.8	percent
Qwen3.8-27B	#3	qwen-3-8-27b	Alibaba Cloud	65.5	percent
GPT-5.4	#4	gpt-5-4	OpenAI	65.4	percent
Muse Spark	#5	muse-spark	Meta	64.7	percent
Qwen3.6 27B	#6	qwen3-6-27b	Alibaba Cloud	62.5	percent
Grok 4.20	#7	grok-4-20	xAI	54.1	percent
Claude Opus 4.6	#8	claude-opus-4-6	Anthropic	51.6	percent

ERQA leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Qwen3.8-Flash-Next Qwen/Qwen3.8-Flash-Next	Alibaba Cloud	open	Public reference	72.3%
#2	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	closed	Source carrier	69.8%
#3	Qwen3.8-27B Qwen/Qwen3.8-27B	Alibaba Cloud	open	Public reference	65.5%
#4	GPT-5.4 gpt-5.4	OpenAI	closed	Source carrier	65.4%
#5	Muse Spark muse-spark	Meta	closed	Source carrier	64.7%
#6	Qwen3.6 27B qwen3.6-27b	Alibaba Cloud	open	Source carrier	62.5%
#7	Grok 4.20 grok-4.20	xAI	closed	Source carrier	54.1%
#8	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	Source carrier	51.6%

Rank

Model

Provider

License

Evidence use

Score

Qwen3.8-Flash-Next

Qwen/Qwen3.8-Flash-Next

Alibaba Cloud

open

Public reference

72.3%

Qwen3.7-Plus

qwen3.7-plus

Alibaba Cloud

closed

Source carrier

69.8%

Qwen3.8-27B

Qwen/Qwen3.8-27B

Alibaba Cloud

open

Public reference

65.5%

GPT-5.4

gpt-5.4

OpenAI

closed

Source carrier

65.4%

Muse Spark

muse-spark

Every distinct dated source-published row for ERQA, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Qwen3.8-Flash-Next (xhigh default thinking configuration) qwen-3-8-flash-next-xhigh	Qwen3.8-Flash-Next	72.3%	2026system:qwen3-8-flash-next:erqa	Public reference · not scored source-checked	Qwen3.8-Flash-Next current official launch page ↗Observed 2026-08-26 · checked 2026-08-29
Qwen3.7-Plus (provider-published configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table qwen-3-7-plus-unspecified	Qwen3.7-Plus	69.8%	2026system:qwen3-8-flash-next-comparison:cell:vision:erqa:2	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
Qwen3.8-27B (xhigh default thinking configuration) as published in Qwen's Qwen3.8-Flash-Next comparison table qwen-3-8-27b-xhigh	Qwen3.8-27B	65.5%	2026system:qwen3-8-flash-next-comparison:cell:vision:erqa:1	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
Claude Opus 4.6 (Max) as published in Qwen's Qwen3.8-Flash-Next comparison table claude-opus-4-6-max	Claude Opus 4.6	40.8%	2026system:qwen3-8-flash-next-comparison:cell:vision:erqa:3	Relative comparison · not standard direct provider-reported	Qwen3.8-Flash-Next launch and official provider evaluations ↗Observed 2026-08-26 · checked 2026-08-26
Qwen3.7-Plus as published by Qwen qwen-3-7-plus-unspecified	Qwen3.7-Plus	69.8%	2026Qwen3.8-27B official VL table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Qwen3.8-27B (xhigh) qwen-3-8-27b-xhigh	Qwen3.8-27B	65.5%	2026Qwen3.8-27B official VL table	Public reference · not scored provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Qwen3.6-27B as published by Qwen qwen3-6-27b-default	Qwen3.6 27B	62.5%	2026Qwen3.8-27B official VL table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Claude Opus 4.6 Max as published by Qwen claude-opus-4-6-max	Claude Opus 4.6	40.8%	2026Qwen3.8-27B official VL table	Relative comparison · not standard direct provider-reported	Qwen3.8-27B official model card ↗Observed 2026-08-15 · checked 2026-08-15
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	69.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	65.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	64.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	62.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	54.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	51.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	69.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	65.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	64.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	62.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	54.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	51.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Qwen3.7 Plus; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.7-Plus	69.8%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	65.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	64.7%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Qwen3.6-27B; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Qwen3.6 27B	62.5%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	54.1%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.6; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.6	51.6%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21

What does ERQA measure?

A grounded visual reasoning benchmark focused on evidence-based question answering over real images.

Which model scores highest on ERQA?

Qwen3.8-Flash-Next by Alibaba Cloud currently leads with 72.3%.

How many models are evaluated on ERQA?

8 canonical model products have a qualifying score, backed by 26 distinct dated source rows.

Does this affect overall Lumina rank?

No. This benchmark is display-only and does not enter the overall Lumina composite.

Navigation

Leaderboard · 8 models

Qwen3.8-Flash-Next

Qwen3.7-Plus

Qwen3.8-27B

ERQA leaderboard

Crop ERQA leaderboard

Benchmark score table (8 models)

Crop ERQA benchmark table

Crop ERQA published source rows

Compare top models

Definition and scoring

Quick answers

What does ERQA measure?

Which model scores highest on ERQA?

How many models are evaluated on ERQA?

Does this affect overall Lumina rank?

ERQA

Leaderboard · 8 models

Qwen3.8-Flash-Next

Qwen3.7-Plus

Qwen3.8-27B

ERQA leaderboard

Crop ERQA leaderboard

Benchmark score table (8 models)

Crop ERQA benchmark table

Crop ERQA published source rows

Compare top models

Definition and scoring

Quick answers

What does ERQA measure?

Which model scores highest on ERQA?

How many models are evaluated on ERQA?

Does this affect overall Lumina rank?

Leaderboard · 8 models

Qwen3.8-Flash-Next

Qwen3.7-Plus

Qwen3.8-27B

Benchmark score table (8 models)

Compare top models

Definition and scoring

Quick answers

What does ERQA measure?

Which model scores highest on ERQA?

How many models are evaluated on ERQA?

Does this affect overall Lumina rank?

More Multimodal benchmarks

Leaderboard · 8 models

Qwen3.8-Flash-Next

Qwen3.7-Plus

Qwen3.8-27B

Benchmark score table (8 models)

Compare top models

Definition and scoring

Quick answers

What does ERQA measure?

Which model scores highest on ERQA?

How many models are evaluated on ERQA?

Does this affect overall Lumina rank?

More Multimodal benchmarks