HolyGrAIL

All tasks / Models

GPT-6.1 Sol

OpenAI's mid-priced GPT tier, described as near-Astra performance for complex work at lower cost.

  • OpenAI
  • Closed
  • Released 29 Sep 2026

Official page

Where it ranks

  • S
    Chat & assistants
    Models · Cloud · 1,484 on LMArena Text Arena (overall, style control)
  • S
    Search & research
    Models · Cloud · 1,250 on LMArena Search Arena (style control)
  • A
    Writing
    Models · Cloud · 1,458 on LMArena Text Arena: Creative Writing (style control)
  • –
    Documents & OCR
    Models · Cloud · no result on this task’s benchmarks yet
  • A
    Coding
    Models · Cloud · 1,754 on LMArena WebDev / Code Arena (overall)
  • S
    Image understanding
    Models · Cloud · 1,294 on LMArena Vision Arena (overall, style control)
  • –
    Browser & computer agents
    Models · Cloud · no result on this task’s benchmarks yet
  • B
    Workflow automation
    Models · Cloud · 64.9% on AutomationBench-AA (Artificial Analysis)

Facts

Vendor
OpenAI
Family
GPT Sol
Released
29 Sep 2026
Weights
Closed: API and apps only
License
proprietary
API price
$2 in / $10 out per 1M tokens
Access
Paid only
Last verified
18 Oct 2026

Price from developers.openai.com, 18 Oct 2026.

Benchmark results

BenchmarkResultEntry on the leaderboardAs ofSource
Artificial Analysis Intelligence Index52GPT-6.1 Sol (Max)
Raw value from the page's embedded model data; the site displays it rounded to an integer.
18 Oct 2026artificialanalysis.ai
Artificial Analysis MMMU-Pro (independent evaluation)86%GPT-6.1 Sol (Max)
Raw fraction 0.8595 from the model data embedded in https://artificialanalysis.ai/leaderboards/models, converted to percent; rank 6 of 124 entries with an MMMU-Pro result; no last-updated date shown.
19 Oct 2026artificialanalysis.ai
AutomationBench-AA (Artificial Analysis)64.9%GPT-6.1 Sol (Max)
AutomationBench-AA Score (share of each task's objectives completed, a task with any guardrail violation scoring zero); raw value from the page's embedded model data (displayed rounded); rank 15 of 32; Tasks Completed 36.99%, Objectives Completed 88.45%; no last-updated date shown, so as_of = retrieval date.
19 Oct 2026artificialanalysis.ai
LMArena Search Arena (style control)1,250gpt-6.1-sol-max
From LMArena's official leaderboard-dataset (search_style_control, split latest, category overall, published 2026-10-13), which mirrors https://arena.ai/leaderboard/search (site blocks automated fetches). Rank 2 of 42; 5555 votes; 95% CI 1240.9-1258.9.
13 Oct 2026huggingface.co
LMArena Text Arena (overall, style control)1,484gpt-6.1-sol-max
From LMArena's official leaderboard-dataset (text_style_control, split latest, category overall, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text (site blocks automated fetches).
15 Oct 2026huggingface.co
LMArena Text Arena: Creative Writing (style control)1,458gpt-6.1-sol-max
From LMArena's official leaderboard-dataset (text_style_control, split latest, category creative_writing, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text/creative-writing (site blocks automated fetches). Rank 30 of 414; 1419 votes; 95% CI 1441.5-1475.4.
15 Oct 2026huggingface.co
LMArena Vision Arena (overall, style control)1,294gpt-6.1-sol-max
From LMArena's official leaderboard-dataset (vision_style_control, split latest, category overall, published 2026-10-09), which mirrors https://arena.ai/leaderboard/vision (site blocks automated fetches). Rank 8 of 165; 1813 votes; 95% CI 1279.6-1308.5.
9 Oct 2026huggingface.co
LMArena WebDev / Code Arena (overall)1,754GPT-6.1 Sol (Max)
From LMArena's official leaderboard-dataset (webdev = Code Arena, split latest, category overall, published 2026-10-18), which mirrors https://arena.ai/leaderboard/webdev. Rank 4; 3021 votes; 95% CI 1741.3-1767.4.
18 Oct 2026huggingface.co
ParseBench (models)70.5%OpenAI GPT-6.1 Sol (Reasoning Low)
Overall (mean of tables, charts, content faithfulness, semantic formatting, visual grounding) from leaderboard.csv in the ParseBench repo (last leaderboard commit 2026-10-09). Only GPT-6.1 Sol setting listed; GPT-6 Sol rows are the previous version.
9 Oct 2026github.com
SWE-Bench Pro V2 (public set, full)96.1%GPT-6.1-Sol (mini-swe-agent) xhigh
SWE-Bench Pro V2 Full (642 tasks); row added 2026-10-07; 95% CI +/-1.24; harness in the variant name; read from the leaderboard page's embedded data.
19 Oct 2026labs.scale.com

Where you can use it

Earlier versions

  • GPT-6 Sol · 22 Sep 2026
  • GPT-5.6 Sol · 9 Jul 2026

Compare

Alternatives to GPT-6.1 Sol

Sources