Benchmark suiteLatest source data: Oct 18, 2026Checked: October 20, 2026

Reasoning and knowledge rankings.

A clean reasoning surface that starts with available Arena signals and keeps LiveBench, HELM Capabilities, MMLU-Pro, GPQA, and AIME-style math benchmarks visible as planned source integrations.

Ranked models

141

Text, docs, or vision signal

Reasoning variants

69

Models tagged from public names

Top proxy

Claude Fable 5 High

100 reasoning proxy

LiveBench

Planned

Fresh reasoning feed

Reasoning formula

A proxy until dedicated reasoning feeds are wired.

The current index blends Text, Docs, and Vision Arena scores. It is useful for broad comparison, but dedicated LiveBench, GPQA, AIME, MMLU-Pro, and HELM signals should be treated as the next data layer.

Reasoning and knowledge proxy

Higher proxy index is better. Dedicated reasoning benchmarks are source-ready but not silently mixed in yet.

Top 30 of 141

#1
Claude Fable 5 High

claude-fable-5-high

AnthropicProprietary
100reasoning proxy
Text
Not listedText Arena
Docs
Not listedDocs Arena
Vision
1,308Vision Arena rank #1
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#2
Gemini 4 Argon

Gemini 4 Argon (High)

GoogleProprietary
100reasoning proxy
Text
1,524Text Arena rank #1
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#3
Claude Opus 4.6 High

claude-opus-4-6-high

AnthropicProprietary
94reasoning proxy
Text
Not listedText Arena
Docs
1,495Docs Arena rank #2
Vision
1,299Vision Arena rank #3
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#4
Claude Opus 4.7

claude-opus-4-7

AnthropicProprietary
93reasoning proxy
Text
Not listedText Arena
Docs
1,494Docs Arena rank #3
Vision
1,298Vision Arena rank #4
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#5
Claude Opus 4.7 High

claude-opus-4-7-high

AnthropicProprietary
93reasoning proxy
Text
Not listedText Arena
Docs
1,493Docs Arena rank #5
Vision
1,298Vision Arena rank #5
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#6
Claude Opus 4.6

claude-opus-4-6

AnthropicProprietary
92reasoning proxy
Text
Not listedText Arena
Docs
1,494Docs Arena rank #4
Vision
1,294Vision Arena rank #7
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#7
Gemini 3.7 Flash High

gemini-3.7-flash-high

GoogleProprietary
91reasoning proxy
Text
Not listedText Arena
Docs
Not listedDocs Arena
Vision
1,297Vision Arena rank #6
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#8
Claude Fable 5.1 Max

claude-fable-5.1-max

AnthropicProprietary
88reasoning proxy
Text
Not listedText Arena
Docs
1,493Docs Arena rank #6
Vision
1,286Vision Arena rank #17
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#9
Claude Opus 5.5 High

claude-opus-5.5-high

AnthropicProprietary
88reasoning proxy
Text
Not listedText Arena
Docs
Not listedDocs Arena
Vision
1,293Vision Arena rank #11
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#10
Claude Opus 5 High

claude-opus-5-high

AnthropicProprietary
87reasoning proxy
Text
Not listedText Arena
Docs
1,490Docs Arena rank #7
Vision
1,287Vision Arena rank #15
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#11
Claude Fable 5

Claude Fable 5 (High)

AnthropicProprietary
85reasoning proxy
Text
1,504Text Arena rank #4
Docs
1,499Docs Arena rank #1
Vision
Not listedVision Arena
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#12
Gemini 3.8 Flash High

gemini-3.8-flash-high

GoogleProprietary
84reasoning proxy
Text
Not listedText Arena
Docs
Not listedDocs Arena
Vision
1,289Vision Arena rank #14
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#13
Muse Wonder

Muse Wonder (Max)

MetaProprietary
83reasoning proxy
Text
1,509Text Arena rank #2
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#14
Claude Opus 5.5

Claude Opus 5.5 (High)

AnthropicProprietary
81reasoning proxy
Text
1,508Text Arena rank #3
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#15
Muse Spark 1.3 Max

muse-spark-1.3-max

MetaProprietary
77reasoning proxy
Text
Not listedText Arena
Docs
1,468Docs Arena rank #19
Vision
1,290Vision Arena rank #12
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#16
Claude Fable 5.1

Claude Fable 5.1 (Max)

AnthropicProprietary
74reasoning proxy
Text
1,502Text Arena rank #6
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#17
Muse Spark 1.2

Muse Spark 1.2 (xHigh)

MetaProprietary
72reasoning proxy
Text
1,493Text Arena rank #12
Docs
Not listedDocs Arena
Vision
1,293Vision Arena rank #10
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#18
GPT-5.6 Sol xHigh

gpt-5.6-sol-xhigh

OpenAIProprietary
71reasoning proxy
Text
1,485Text Arena rank #21
Docs
1,489Docs Arena rank #8
Vision
1,287Vision Arena rank #16
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#19
Muse Spark 1.3

Muse Spark 1.3 (Max)

MetaProprietary
71reasoning proxy
Text
1,494Text Arena rank #10
Docs
Not listedDocs Arena
Vision
1,286Vision Arena rank #18
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#20
Muse Spark 1.1

Muse Spark 1.1

MetaProprietary
70reasoning proxy
Text
1,490Text Arena rank #13
Docs
1,479Docs Arena rank #12
Vision
1,280Vision Arena rank #28
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#21
Muse Spark

Muse Spark

MetaProprietary
69reasoning proxy
Text
1,489Text Arena rank #16
Docs
1,467Docs Arena rank #20
Vision
1,294Vision Arena rank #9
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#22
Claude Opus 4.6

Claude Opus 4.6

AnthropicProprietary
69reasoning proxy
Text
1,497Text Arena rank #8
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#23
Claude Opus 4.8 High

claude-opus-4-8-high

AnthropicProprietary
68reasoning proxy
Text
1,481Text Arena rank #25
Docs
1,486Docs Arena rank #9
Vision
1,286Vision Arena rank #21
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#24
Gemini 3.8 Flash

Gemini 3.8 Flash (High)

GoogleProprietary
67reasoning proxy
Text
1,496Text Arena rank #9
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#25
Qwen3.8 Max

qwen3.8-max

AlibabaProprietary
66reasoning proxy
Text
1,482Text Arena rank #24
Docs
Not listedDocs Arena
Vision
1,301Vision Arena rank #2
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#26
GPT-6.1 Sol Max

gpt-6.1-sol-max

OpenAIProprietary
66reasoning proxy
Text
1,484Text Arena rank #22
Docs
Not listedDocs Arena
Vision
1,294Vision Arena rank #8
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#27
GPT-5.5 High

gpt-5.5-high

OpenAIProprietary
65reasoning proxy
Text
1,481Text Arena rank #26
Docs
1,480Docs Arena rank #11
Vision
1,281Vision Arena rank #25
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#28
Claude Opus 4.7

Claude Opus 4.7

AnthropicProprietary
65reasoning proxy
Text
1,494Text Arena rank #11
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#29
GPT-5.5

gpt-5.5

OpenAIProprietary
65reasoning proxy
Text
1,477Text Arena rank #31
Docs
1,483Docs Arena rank #10
Vision
1,286Vision Arena rank #20
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#30
Gemini 3.1 Pro Preview

Gemini 3.1 Pro Preview

GoogleProprietary
63reasoning proxy
Text
1,487Text Arena rank #19
Docs
1,459Docs Arena rank #27
Vision
1,280Vision Arena rank #29
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena

Benchmark guide

What the scores mean.

A quick reading key for reasoning and knowledge rankings while dedicated reasoning benchmark feeds are being integrated.

Higher: reasoning proxyLiveBench planned
What does the reasoning proxy measure?

The reasoning proxy currently blends Text, Document, and Vision Arena scores. It is an orientation layer for broad model capability, not a replacement for dedicated reasoning benchmarks like LiveBench, GPQA, AIME, or MMLU-Pro.

Why use Arena scores for reasoning?

Arena scores are live public signals with broad model coverage. They are useful while dedicated reasoning feeds are being wired, but they should be read as preference-based capability signals rather than exam-style accuracy scores.

What will LiveBench add?

LiveBench is designed to update over time and reduce benchmark contamination risk. Once integrated, it can add more direct reasoning, math, data, and coding task signals to this page.

Why are reasoning modes only a signal?

A model name or tag like Thinking, Reasoning, or High can indicate an inference mode, but it is not itself a benchmark score. The page treats those labels as context, not as proof of better reasoning.