Benchmark suiteLatest source data: Jul 16, 2026Checked: July 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

148

Text, docs, or vision signal

Reasoning variants

43

Models tagged from public names

Top proxy

Claude Fable 5

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 148

#1
Claude Fable 5

claude-fable-5

AnthropicProprietary
100reasoning proxy
Text
1,507Text Arena rank #1
Docs
1,509Docs Arena rank #1
Vision
1,318Vision Arena rank #1
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#2
Claude Opus 4.7 Thinking

claude-opus-4-7-thinking

AnthropicProprietary
94reasoning proxy
Text
1,502Text Arena rank #3
Docs
1,503Docs Arena rank #2
Vision
1,304Vision Arena rank #2
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#3
Claude Opus 4.6 Thinking

claude-opus-4-6-thinking

AnthropicProprietary
92reasoning proxy
Text
1,504Text Arena rank #2
Docs
1,494Docs Arena rank #4
Vision
1,299Vision Arena rank #3
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#4
Claude Opus 4.7

claude-opus-4-7

AnthropicProprietary
88reasoning proxy
Text
1,494Text Arena rank #5
Docs
1,500Docs Arena rank #3
Vision
1,299Vision Arena rank #4
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#5
Claude Opus 4.6

claude-opus-4-6

AnthropicProprietary
88reasoning proxy
Text
1,498Text Arena rank #4
Docs
1,491Docs Arena rank #5
Vision
1,298Vision Arena rank #5
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#6
Muse Spark 1.1

muse-spark-1.1

MetaProprietary
84reasoning proxy
Text
1,493Text Arena rank #6
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#7
Claude Opus 4.8 Thinking

claude-opus-4-8-thinking

AnthropicProprietary
78reasoning proxy
Text
1,483Text Arena rank #12
Docs
1,487Docs Arena rank #6
Vision
1,286Vision Arena rank #9
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#8
Muse Spark

muse-spark

MetaProprietary
76reasoning proxy
Text
1,487Text Arena rank #7
Docs
1,469Docs Arena rank #13
Vision
1,295Vision Arena rank #6
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#9
Kimi K3

kimi-k3

Moonshot AIProprietary
76reasoning proxy
Text
1,486Text Arena rank #10
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#10
GPT-5.6 Sol xHigh

gpt-5.6-sol-xhigh

OpenAIProprietary
76reasoning proxy
Text
1,486Text Arena rank #8
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#11
GPT-5.5 High

gpt-5.5-high

OpenAIProprietary
75reasoning proxy
Text
1,482Text Arena rank #13
Docs
1,480Docs Arena rank #8
Vision
1,286Vision Arena rank #8
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#12
GPT-5.4 High

gpt-5.4-high

OpenAIProprietary
72reasoning proxy
Text
1,478Text Arena rank #14
Docs
Not listedDocs Arena
Vision
1,281Vision Arena rank #13
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#13
Claude Opus 4.8

claude-opus-4-8

AnthropicProprietary
72reasoning proxy
Text
1,476Text Arena rank #16
Docs
1,482Docs Arena rank #7
Vision
1,282Vision Arena rank #12
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#14
Gemini 3 Pro

gemini-3-pro

GoogleProprietary
71reasoning proxy
Text
1,486Text Arena rank #9
Docs
1,452Docs Arena rank #19
Vision
1,289Vision Arena rank #7
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#15
GPT-5.2 Chat Latest 20260210

gpt-5.2-chat-latest-20260210

OpenAIProprietary
70reasoning proxy
Text
1,476Text Arena rank #17
Docs
Not listedDocs Arena
Vision
1,279Vision Arena rank #14
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#16
Gemini 3.1 Pro Preview

gemini-3.1-pro-preview

GoogleProprietary
70reasoning proxy
Text
1,485Text Arena rank #11
Docs
1,456Docs Arena rank #17
Vision
1,278Vision Arena rank #15
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#17
Gemini 3.5 Flash High

gemini-3.5-flash-high

GoogleProprietary
70reasoning proxy
Text
1,476Text Arena rank #15
Docs
Not listedDocs Arena
Vision
1,276Vision Arena rank #18
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#18
GPT-5.5

gpt-5.5

OpenAIProprietary
70reasoning proxy
Text
1,475Text Arena rank #19
Docs
1,474Docs Arena rank #11
Vision
1,285Vision Arena rank #11
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#19
Claude Sonnet 4.6

claude-sonnet-4-6

AnthropicProprietary
69reasoning proxy
Text
1,472Text Arena rank #26
Docs
1,480Docs Arena rank #9
Vision
1,276Vision Arena rank #17
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#20
Gemini 3.5 Flash Medium

gemini-3.5-flash-medium

GoogleProprietary
66reasoning proxy
Text
1,474Text Arena rank #20
Docs
1,460Docs Arena rank #15
Vision
1,286Vision Arena rank #10
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#21
Qwen3.7 Max Preview

qwen3.7-max-preview

AlibabaProprietary
64reasoning proxy
Text
1,475Text Arena rank #18
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#22
GPT-5.5 Instant

gpt-5.5-instant

OpenAIProprietary
64reasoning proxy
Text
1,473Text Arena rank #23
Docs
1,457Docs Arena rank #16
Vision
1,276Vision Arena rank #19
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#23
GPT-5.4

gpt-5.4

OpenAIProprietary
63reasoning proxy
Text
1,466Text Arena rank #32
Docs
1,471Docs Arena rank #12
Vision
1,278Vision Arena rank #16
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#24
Grok 4.20 Beta 1

grok-4.20-beta1

xAIProprietary
63reasoning proxy
Text
1,474Text Arena rank #21
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#25
Claude Opus 4.5 20251101 Thinking 32K

claude-opus-4-5-20251101-thinking-32k

AnthropicProprietary
62reasoning proxy
Text
1,473Text Arena rank #25
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#26
Grok 4.20 Multi Agent Beta 0309

grok-4.20-multi-agent-beta-0309

xAIProprietary
61reasoning proxy
Text
1,470Text Arena rank #28
Docs
Not listedDocs Arena
Vision
1,250Vision Arena rank #29
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#27
Claude Sonnet 5 High

claude-sonnet-5-high

AnthropicProprietary
61reasoning proxy
Text
1,462Text Arena rank #37
Docs
1,477Docs Arena rank #10
Vision
1,264Vision Arena rank #22
Mode
Reasoningmode signal
Text ArenaDocs ArenaVision Arena
#28
GLM 5.1

glm-5.1

Z.aiOpen weights
60reasoning proxy
Text
1,471Text Arena rank #27
Docs
Not listedDocs Arena
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#29
Claude Opus 4.5 20251101

claude-opus-4-5-20251101

AnthropicProprietary
57reasoning proxy
Text
1,469Text Arena rank #29
Docs
1,463Docs Arena rank #14
Vision
Not listedVision Arena
Mode
Standardmode signal
Text ArenaDocs ArenaVision Arena
#30
Grok 4.20 Beta 0309 Reasoning

grok-4.20-beta-0309-reasoning

xAIProprietary
57reasoning proxy
Text
1,473Text Arena rank #24
Docs
1,441Docs Arena rank #24
Vision
1,253Vision Arena rank #28
Mode
Reasoningmode 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.