Models  ›  Text  ›  GPT 6.1 Sol

OPENAI GPT-6 SERIES MODEL 2026

GPT-6.1 Sol

OPEN AI GPT 6 SERIES MODEL 2026

GPT-6.1 Sol is OpenAI’s high-end model below the flagship Astra

Text generation
Extended Reasoning
Long-horizon agentic tool use
CONTEXT WINDOW
1,050,000 tokens
MAXIMUM OUTPUT
128,000 tokens
BASE PROVIDER API INPUT
$2.00 / 1M tokens

Starting provider API rate; context tiers apply

PUBLISHED EVIDENCE
5 benchmarks, 0 Arena results
OVERVIEW

About GPT-6.1 Sol

See what GPT-6.1 Sol does best, then review its limits, benchmarks, and pricing below

GPT-6.1 Sol handles demanding reasoning, coding, and everyday professional work

Text
Reasoning
PUBLISHED EVALUATIONS

Sol Benchmarks

Every GPT-6.1 Sol result below comes from a named source and is labeled by protocol, so each score reflects how it was measured

BenchmarkScoreEvaluation detailsSource
AutomationBench-AA
Tests whether GPT-6.1 Sol completes multi-step workflows across simulated SaaS apps in six domains.
FILL — e.g. 69%
Guardrail-aware score
Version: AutomationBench-AA 2026 • Split: 657 tasks across Finance, HR, Marketing, Operations, Sales, and Support • Harness: Artificial Analysis independent implementation • Attempts: One evaluated run per task • Evaluator: Programmatic end-state grader with guardrail checks • Reasoning effort: max Methodology
Artificial Analysis
Independently reproduced
DeepSWE 1.1
Measures GPT-6.1 Sol on 113 long-horizon software-engineering tasks graded by hand-written tests.
FILL — e.g. 74%±3
Pass@1
Version: DeepSWE 1.1 • Split: 113 tasks across 91 repositories • Harness: mini-swe-agent • Coverage: five languages • Confidence interval: ±3 Methodology
DataCurve
Benchmark owner
GPQA Diamond
Tests GPT-6.1 Sol on 198 Google-proof graduate-level questions in biology, chemistry, and physics to probe expert-level scientific reasoning.
FILL — e.g. 96.0%
Accuracy
Version: GPQA Diamond 198 • Split: 198 Diamond questions • Harness: FILL — evaluation table or harness used • Reasoning effort: max Methodology
OpenAI
Provider-reported
Humanity's Last Exam
Tests GPT-6.1 Sol on 2,500 expert-level questions across dozens of academic fields, with tool use enabled.
FILL — e.g. 57.2%
Accuracy
Version: Humanity's Last Exam • Protocol: Tool-assisted, reported separately from no-tools results • Harness: FILL — evaluation table or harness used • Score basis: Maximum result at any reasoning effort Methodology
OpenAI
Provider-reported
Terminal-Bench 4.0
Tests GPT-6.1 Sol on complex terminal-based tasks spanning software engineering, system configuration, and data analysis, rewarding long-horizon tool use and error recovery.
FILL — e.g. 59%
Independent task success rate
Version: Terminal-Bench 4.0 • Harness: Artificial Analysis evaluation • Index: Intelligence Index component • Reasoning effort: max Methodology
Artificial Analysis
Independently reproduced
HUMAN PREFERENCE

Sol Arena

Arena rankings capture how people vote between anonymous responses. GPT-6.1 Sol’s human preference standing is reported per leaderboard, since results can vary widely by task

FILL — 18-word shape: “GPT-6.1 Sol ranked [n] on Arena’s [leaderboard] in [month], scoring [score] from [votes] votes”

TECHNICAL REFERENCE

Sol Specifications and Pricing

GPT-6.1 Sol offers a-token context window, output tokens, text input, and adjustable reasoning effort, with full API pricing and tiers shown below

Model
ProviderOpenAI
Release dateSep 3, 2026
Catalog addedSep 22, 2026
Context window1,050,000 tokens
Maximum output128,000 tokens
Knowledge cutoff2026-04-30
LicenseProprietary
Capabilities and API
Input typesText
Output typesText
ReasoningSupported
Provider endpoint tool parametersAccepts tool parameters
Base provider API input$10.00 / 1M tokens
Base provider cached input (read)$1.00 / 1M tokens
Base provider API output$50.00 / 1M tokens
Provider tier above [threshold] prompt tokensInput $X / 1M • Output $Y / 1M • Cached input $Z / 1M

Provider token rates are context-dependent. The base rates apply below the listed prompt thresholds; the matching tier applies at or above each threshold. These are reference provider API rates, not Pyxa AI checkout charges — access in Pyxa AI follows the allowances of your Pyxa AI plan.

PROVENANCE

Sol Sources and Updates

Each GPT-6.1 Sol benchmark and specification links to its original source, so you can check where every figure comes from

AutomationBench-AA Leaderboard
Artificial Analysis · Retrieved Nov 5, 2026
DeepSWE 1.1 Benchmark
DataCurve · Retrieved Nov 5, 2026
GPQA Diamond Benchmark
OpenAI · Retrieved Nov 5, 2026
Humanity's Last Exam Benchmark
OpenAI · Retrieved Nov 5, 2026
Terminal-Bench 4.0 Leaderboard
Artificial Analysis · Retrieved Nov 5, 2026
GPT-6 Sol Model Documentation
OpenAI · Retrieved Nov 5, 2026
Evidence last updated Nov 5, 2026. Catalog record updated Nov 5, 2026. Catalog-added dates describe when a model entered the catalog, not necessarily its public release date.

Try GPT-6.1 Sol in Pyxa AI

Put GPT-6.1 Sol to work on your reasoning, coding, and everyday tasks