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
128
Docs, search, or text signal
Docs coverage
38
Document Arena rows
Search coverage
34
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 128
| Rank | Model | Index | Docs | Search | Text | Context | Sources |
|---|---|---|---|---|---|---|---|
| #1 | 98context proxy | Not listedDocs Arena | Not listedSearch Arena | 1,505Text Arena rank #2 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #2 | 96context proxy | 1,507Docs Arena rank #1 | 1,229Search Arena rank #2 | 1,507Text Arena rank #1 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #3 | 94context proxy | Not listedDocs Arena | Not listedSearch Arena | 1,502Text Arena rank #3 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #4 | 90context proxy | Not listedDocs Arena | Not listedSearch Arena | 1,499Text Arena rank #4 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #5 | 90context proxy | 1,496Docs Arena rank #3 | Not listedSearch Arena | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #6 | 89context proxy | 1,495Docs Arena rank #4 | Not listedSearch Arena | 1,498Text Arena rank #5 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #7 | 87context proxy | 1,493Docs Arena rank #5 | Not listedSearch Arena | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #8 | 85context proxy | 1,497Docs Arena rank #2 | 1,212Search Arena rank #5 | 1,494Text Arena rank #6 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #9 | 85context proxy | 1,493Docs Arena rank #6 | Not listedSearch Arena | 1,492Text Arena rank #7 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #10 | 84context proxy | Not listedDocs Arena | 1,224Search Arena rank #3 | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #11 | 83context proxy | Not listedDocs Arena | 1,223Search Arena rank #4 | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #12 | 81context proxy | 1,487Docs Arena rank #7 | Not listedSearch Arena | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #13 | 81context proxy | 1,481Docs Arena rank #11 | 1,256Search Arena rank #1 | 1,483Text Arena rank #16 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #14 | 80context proxy | Not listedDocs Arena | Not listedSearch Arena | 1,491Text Arena rank #9 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #15 | 79context proxy | 1,483Docs Arena rank #8 | Not listedSearch Arena | 1,492Text Arena rank #8 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #16 | 78context proxy | Not listedDocs Arena | Not listedSearch Arena | 1,489Text Arena rank #10 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #17 | 76context proxy | Not listedDocs Arena | Not listedSearch Arena | 1,488Text Arena rank #12 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #18 | 75context proxy | Not listedDocs Arena | 1,208Search Arena rank #6 | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #19 | 72context proxy | Not listedDocs Arena | 1,201Search Arena rank #9 | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #20 | 72context proxy | Not listedDocs Arena | 1,201Search Arena rank #10 | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #21 | 71context proxy | 1,478Docs Arena rank #13 | Not listedSearch Arena | 1,482Text Arena rank #17 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #22 | 70context proxy | Not listedDocs Arena | Not listedSearch Arena | 1,483Text Arena rank #15 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #23 | 70context proxy | 1,482Docs Arena rank #10 | 1,195Search Arena rank #13 | 1,473Text Arena rank #31 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #24 | 69context proxy | Not listedDocs Arena | 1,196Search Arena rank #12 | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #25 | 69context proxy | Not listedDocs Arena | Not listedSearch Arena | 1,482Text Arena rank #18 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #26 | 69context proxy | Not listedDocs Arena | 1,195Search Arena rank #14 | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #27 | 68context proxy | Not listedDocs Arena | 1,194Search Arena rank #15 | Not listedText Arena | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #28 | 67context proxy | 1,475Docs Arena rank #15 | Not listedSearch Arena | 1,477Text Arena rank #22 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #29 | 67context proxy | 1,482Docs Arena rank #9 | Not listedSearch Arena | 1,466Text Arena rank #41 | HELM pendinglong-context feed | Docs ArenaSearch ArenaText Arena | |
| #30 | 67context proxy | Not listedDocs Arena | 1,192Search Arena rank #18 | Not listedText Arena | 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.
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.
