Model results · OpenAI

GPT-6.1 Sol

Every published VulcanBench measurement of GPT-6.1 Sol, newest first: combined score, Code quality, runtime, tokens and API-equivalent cost at every effort level, graded by deterministic hidden tests on real tasks. Measured through the Codex CLI on a ChatGPT subscription. GPT-6.1 Sol is a newer model than GPT-6 Sol and GPT-5.6 Sol, which have their own pages.

Suitev4

GPT-6.1 Sol across every effort level

In Codex across five effort levels: combined score 86.22 at Low, 86.44 at Medium, 88.23 at High, 88.78 at Extra-high and 88.35 at Max. Tasks passed 20, 22, 23, 23 and 23 of 23, with every hidden test passed from High up; Code quality 70.25 to 74.99; 6.9 to 15.4 minutes per task. Every cell is judged on 23 runs.

October 1, 2026 · 23 tasks · 5 effort levels · 115 runs of GPT-6.1 Sol · VulcanBench Frontier v4 · Code quality protocol v3.18

→

Suitev4

Cost and tokens across effort levels

$0.31 to $0.46 per task at list API rates and 0.91M to 1.62M raw tokens per task across the five effort levels; the whole sweep prices at $44.33 for 115 runs, $0.39 per task.

October 1, 2026 · API-equivalent estimates at list rates, solver inference only · VulcanBench Frontier v4

→

Suitev4

Three generations of Sol

GPT-6.1 Sol leads GPT-6 Sol and GPT-5.6 Sol at every effort level on combined score, tasks passed and cost per task, at 14% to 30% of GPT-6 Sol's cost per task. On the Frontier v4 board its best level ranks 13th of 49 columns, below Fable 5.1, Opus 5.5 and GPT-6 Astra, mainly on Code quality.

October 1, 2026 · comparison card · VulcanBench Frontier v4

→

How would GPT-6.1 Sol do on your stack?

Public results answer "which model is stronger on this suite." Whether GPT-6.1 Sol belongs in your routing table depends on your languages, your codebases, and your task mix, that gets measured, not guessed. Grading rules are in the methodology.

Optimize your model routing All reports