Benchmark suiteLatest source data: Jul 16, 2026Checked: July 20, 2026

Long context rankings.

A focused page for document-heavy and retrieval-heavy work. The current view uses Docs, Search, and Text Arena signals while HELM Long Context and context-window metadata are prepared as the next data layer.

Ranked models

125

Docs, search, or text signal

Docs coverage

32

Document Arena rows

Search coverage

32

Search Arena rows

HELM context

Planned

Long-context leaderboard feed

Context formula

Document skill first, retrieval support second.

The context proxy blends Document Arena, Search Arena, and Text Arena scores. It does not claim a context-window size or true long-context pass rate until a dedicated source row is available.

Long-context proxy

Higher proxy index is better. HELM Long Context is shown as a planned source rather than a hidden assumption.

Top 30 of 125

#1
Claude Fable 5

claude-fable-5

AnthropicProprietary
100context proxy
Docs
1,509Docs Arena rank #1
Search
1,229Search Arena rank #1
Text
1,507Text Arena rank #1
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#2
Claude Opus 4.6 Search

claude-opus-4-6-search

AnthropicProprietary
97context proxy
Docs
Not listedDocs Arena
Search
1,224Search Arena rank #2
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#3
GPT-5.5 Search

gpt-5.5-search

OpenAIProprietary
96context proxy
Docs
Not listedDocs Arena
Search
1,222Search Arena rank #3
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#4
Claude Opus 4.7 Thinking

claude-opus-4-7-thinking

AnthropicProprietary
94context proxy
Docs
1,503Docs Arena rank #2
Search
Not listedSearch Arena
Text
1,502Text Arena rank #3
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#5
Claude Opus 4.7

claude-opus-4-7

AnthropicProprietary
90context proxy
Docs
1,500Docs Arena rank #3
Search
1,213Search Arena rank #4
Text
1,494Text Arena rank #5
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#6
Claude Opus 4.6 Thinking

claude-opus-4-6-thinking

AnthropicProprietary
90context proxy
Docs
1,494Docs Arena rank #4
Search
Not listedSearch Arena
Text
1,504Text Arena rank #2
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#7
Gemini 3.1 Pro Grounding

gemini-3.1-pro-grounding

GoogleProprietary
87context proxy
Docs
Not listedDocs Arena
Search
1,208Search Arena rank #5
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#8
Claude Opus 4.6

claude-opus-4-6

AnthropicProprietary
85context proxy
Docs
1,491Docs Arena rank #5
Search
Not listedSearch Arena
Text
1,498Text Arena rank #4
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#9
Muse Spark 1.1

muse-spark-1.1

MetaProprietary
84context proxy
Docs
Not listedDocs Arena
Search
Not listedSearch Arena
Text
1,493Text Arena rank #6
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#10
Gemini 3 Pro Grounding

gemini-3-pro-grounding

GoogleProprietary
83context proxy
Docs
Not listedDocs Arena
Search
1,201Search Arena rank #7
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#11
Claude Sonnet 4.6 Search

claude-sonnet-4-6-search

AnthropicProprietary
83context proxy
Docs
Not listedDocs Arena
Search
1,200Search Arena rank #8
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#12
Claude Sonnet 5 Search

claude-sonnet-5-search

AnthropicProprietary
80context proxy
Docs
Not listedDocs Arena
Search
1,196Search Arena rank #11
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#13
Grok 4.1 Fast Search

grok-4-1-fast-search

xAIProprietary
79context proxy
Docs
Not listedDocs Arena
Search
1,194Search Arena rank #12
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#14
GPT-5.4 Search

gpt-5.4-search

OpenAIProprietary
78context proxy
Docs
Not listedDocs Arena
Search
1,193Search Arena rank #14
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#15
Claude Opus 4.8 Thinking

claude-opus-4-8-thinking

AnthropicProprietary
77context proxy
Docs
1,487Docs Arena rank #6
Search
Not listedSearch Arena
Text
1,483Text Arena rank #12
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#16
O3 Search

o3-search

OpenAIProprietary
77context proxy
Docs
Not listedDocs Arena
Search
1,191Search Arena rank #15
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#17
Gemini 3 Flash Grounding

gemini-3-flash-grounding

GoogleProprietary
77context proxy
Docs
Not listedDocs Arena
Search
1,190Search Arena rank #17
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#18
Kimi K3

kimi-k3

Moonshot AIProprietary
76context proxy
Docs
Not listedDocs Arena
Search
Not listedSearch Arena
Text
1,486Text Arena rank #10
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#19
GPT-5.6 Sol xHigh

gpt-5.6-sol-xhigh

OpenAIProprietary
76context proxy
Docs
Not listedDocs Arena
Search
Not listedSearch Arena
Text
1,486Text Arena rank #8
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#20
Claude Opus 4.5 Search

claude-opus-4-5-search

AnthropicProprietary
75context proxy
Docs
Not listedDocs Arena
Search
1,187Search Arena rank #18
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#21
Claude Opus 4.8

claude-opus-4-8

AnthropicProprietary
74context proxy
Docs
1,482Docs Arena rank #7
Search
1,197Search Arena rank #10
Text
1,476Text Arena rank #16
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#22
Grok 4.20 Multi Agent Beta 0309

grok-4.20-multi-agent-beta-0309

xAIProprietary
74context proxy
Docs
Not listedDocs Arena
Search
1,207Search Arena rank #6
Text
1,470Text Arena rank #28
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#23
GPT-5.1 Search

gpt-5.1-search

OpenAIProprietary
73context proxy
Docs
Not listedDocs Arena
Search
1,184Search Arena rank #19
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#24
Grok 4.20 Beta 1

grok-4.20-beta1

xAIProprietary
73context proxy
Docs
Not listedDocs Arena
Search
1,198Search Arena rank #9
Text
1,474Text Arena rank #21
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#25
GPT-5.5 High

gpt-5.5-high

OpenAIProprietary
73context proxy
Docs
1,480Docs Arena rank #8
Search
Not listedSearch Arena
Text
1,482Text Arena rank #13
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#26
GPT-5 Search

gpt-5-search

OpenAIProprietary
72context proxy
Docs
Not listedDocs Arena
Search
1,183Search Arena rank #20
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#27
Claude Sonnet 4.5 Search

claude-sonnet-4-5-search

AnthropicProprietary
69context proxy
Docs
Not listedDocs Arena
Search
1,177Search Arena rank #21
Text
Not listedText Arena
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#28
Claude Sonnet 4.6

claude-sonnet-4-6

AnthropicProprietary
69context proxy
Docs
1,480Docs Arena rank #9
Search
Not listedSearch Arena
Text
1,472Text Arena rank #26
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#29
ERNIE 5.1

ernie-5.1

BaiduProprietary
68context proxy
Docs
Not listedDocs Arena
Search
1,193Search Arena rank #13
Text
1,468Text Arena rank #31
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena
#30
Muse Spark

muse-spark

MetaProprietary
68context proxy
Docs
1,469Docs Arena rank #13
Search
Not listedSearch Arena
Text
1,487Text Arena rank #7
Context
HELM pendinglong-context feed
Docs ArenaSearch ArenaText Arena

Benchmark guide

What the scores mean.

A quick reading key for long-context rankings, document-heavy work, retrieval signals, and planned context-window data.

Higher: context proxyHELM feed planned
What does the long-context proxy measure?

The long-context proxy blends Document Arena, Search Arena, and Text Arena scores. It is meant to surface models that appear strong on document-heavy and retrieval-heavy work while dedicated long-context benchmarks are being wired.

Does this measure context window size?

Not yet. Context window size and max output tokens need a reliable source before they become table fields. This ranking focuses on public task performance, not advertised token limits.

Why is HELM Long Context listed as planned?

HELM Long Context is the right benchmark family for more direct long-context evaluation, but it is not mixed into the score until the feed is integrated and matched cleanly to model rows.

How should I read document and search scores?

Document scores point toward long-form reading and file-style tasks. Search scores point toward retrieval-style workflows. Both are useful signals, but neither proves that a model can reliably use every token in a huge context window.