---
type: "raw_signal"
id: "gpt-6-sol-now-leads-deepsecbench-s-pareto-803739"
slug: "gpt-6-sol-now-leads-deepsecbench-s-pareto-803739"
title: "GPT-6 Sol now leads DeepSecBench's Pareto frontier: • Highest benchmark score • 80% lower run cost than the runner-up (GPT-6 Astra) https://t.co/ZmDC4NmjSQ https://t.co/soK5ZM0RdD"
description: "Vercel AI Gateway's DeepSecBench leaderboard lists GPT-6 Sol xhigh first with a 40.91 score, 35.8% recall, 96.0% precision, four false positives, and a $12.76 run cost using the Codex harness."
retrieval_nugget: "Vercel AI Gateway's DeepSecBench leaderboard lists GPT-6 Sol xhigh first with a 40.91 score, 35.8% recall, 96.0% precision, four false positives, and a $12.76 run cost using the Codex harness."
observed_at: "2026-09-25"
record_date: "2026-09-25"
date_kind: "observed_at"
why_it_matters: "Source: https://vercel.com/ai-gateway/leaderboards/deepsecbench"
novelty: "notable"
verification_level: "source-linked"
signal_type: "research"
source_platform: "vercel.com"
topics: []
entities: []
related_patterns: []
source_url: "https://vercel.com/ai-gateway/leaderboards/deepsecbench"
source_urls: ["https://vercel.com/ai-gateway/leaderboards/deepsecbench"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:gpt-6-sol-now-leads-deepsecbench-s-pareto-803739"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/signals/gpt-6-sol-now-leads-deepsecbench-s-pareto-803739/","markdown":"https://newruntime.com/signals/gpt-6-sol-now-leads-deepsecbench-s-pareto-803739.md","json":"https://newruntime.com/signals/gpt-6-sol-now-leads-deepsecbench-s-pareto-803739.json"}
---

# GPT-6 Sol now leads DeepSecBench's Pareto frontier: • Highest benchmark score • 80% lower run cost than the runner-up (GPT-6 Astra) https://t.co/ZmDC4NmjSQ https://t.co/soK5ZM0RdD

## Retrieval answer

Vercel AI Gateway's DeepSecBench leaderboard lists GPT-6 Sol xhigh first with a 40.91 score, 35.8% recall, 96.0% precision, four false positives, and a $12.76 run cost using the Codex harness.

Vercel AI Gateway's DeepSecBench leaderboard lists GPT-6 Sol xhigh first with a 40.91 score, 35.8% recall, 96.0% precision, four false positives, and a $12.76 run cost using the Codex harness. GPT-6 Astra xhigh is second at 37.79 and $63.70. On this recorded benchmark configuration, Sol achieved the higher score at roughly 80% lower listed run cost. The comparison is a benchmark-specific operating signal, not a general security or model-quality ranking.

Source: https://vercel.com/ai-gateway/leaderboards/deepsecbench

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)