Skip to content
Leaderboard / OpenAI

GPT-6 Sol

OpenAI Max CLOSED Proprietary
Verdict

GPT-6 Sol is a strong all-rounder: #43 of 406 by SIscope score. Strongest at math, weaker at chat quality. On the pricey side: costs more than 77% of priced models.

SIscope score
80
#43 of 406
Human-vote rank
#159
of 419 · score 1397
Price per 1M tokens, blended
$4
$2 in · $10 out
Context window
1.1M
1,050,000 tokens

Shape

Each spoke is a category, 0 at the centre, 100 at the edge.

CodingAgentsReasoningMathChatInstructions

Category scores

Percentiles: 90 means better than about 90% of tested models.

Coding84
Agents & tools80
Reasoning & knowledge84
Math84
Chat quality62
Following instructions75

What goes into each category · Compare with DeepSeek V4.1 Flash

Human-vote score over time

Text arena, recent updates.

139013981405Oct 20: 1398.1 (rank #157)Oct 22: 1398.2 (rank #157)Oct 28: 1397.4 (rank #159)Oct 20Oct 28

Coding votes over time

WebDev arena, recent updates.

167016801690Oct 16: 1682.7 (rank #8)Oct 18: 1682.4 (rank #8)Oct 19: 1683.1 (rank #9)Oct 22: 1681.6 (rank #9)Oct 23: 1680.8 (rank #10)Oct 27: 1677.4 (rank #11)Oct 28: 1678.2 (rank #12)Oct 16Oct 28

Independent test runs

Run by Epoch AI on the same harness for every model. Bars show the share of questions answered correctly.

GPQA Diamond94%
PhD-level science multiple choice questions
Better than 97% of 333 tested models. Run Sep 22, 2026.
SimpleQA Verified61%
Short factual questions; rewards not making things up
Better than 86% of 100 tested models. Run Sep 22, 2026.
OTIS Mock AIME 2024-25100%
Competition maths problems in the style of AIME
Better than 96% of 313 tested models. Run Sep 22, 2026.
FrontierMath Tiers 1-390%
Very hard, unpublished research-style maths problems
Better than 96% of 122 tested models. Run Sep 22, 2026.

Human-vote arenas

People compare two anonymous answers and pick the better one. Scores work like chess ratings.

ArenaRankScoreRange (95%)VotesUpdated
Text arena
Everyday chat, blind human votes
#159 of 41913971392–140315,807Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#12 of 15216781669–16885,714Oct 28, 2026
Vision arena
Questions about images, blind human votes
#74 of 16712401229–12513,221Oct 19, 2026
Agent arena
Real multi-step agent sessions
#12 of 577.0%5.4–8.5%12,478Nov 2, 2026

Source: Arena (formerly LMArena) leaderboard dataset, CC BY 4.0. Fetched Nov 4, 2026.

Price

Input
$2
per 1M tokens
Output
$10
per 1M tokens
Blended
$4
3 in : 1 out, per 1M
Context window
1.1M
max tokens per request
Speed
Coming soon
our own tests

List price from the OpenRouter models API (openai/gpt-6-sol), Nov 4, 2026. Providers can charge differently.