signalAI热榜2026-09-24
Arena Test: GPT-6 Sol (Max) Ranks 4th in Code Arena: WebDev with a Score of 1689
GPT-6 Sol (Max) scored 1689 in Code Arena: WebDev, ranking 4th, and reshaped the Pareto frontier at a price of $8/M tokens (mixed input/output). Its performance is comparable to Claude Opus 5 (Max) but at less than half the cost, while lagging GPT-6 Astra (Max) by two positions at one-fifth the cost. It improved +72 points over GPT-5.6 Sol (xHigh), with gains across all categories, including simulation, content creation, and games.
- for who
- Developers and AI model users seeking high performance at lower cost
- why now
- GPT-6 Sol Max launches at $8/M, matching Opus 5 at half cost, ranking fourth.
- what changes
- High-end coding and creation tasks become affordable with a near-top-tier model
- to do
- Evaluate GPT-6 Sol (Max) for large-scale workloads and compare it with other models
key points
- 1689 points, 4th in Code Arena: WebDev
- $8/M tokens, half the cost of Claude Opus 5
- +72 points over GPT-5.6 Sol, top gains in all categories
#gpt-6#benchmark#coding ability#pricing
score
score 8 out of 10. 0-10: how dense the facts are, multiplied by how much you can do with them after reading. 8+ means the topic's evidence bar is met: benchmarks and availability for a new model, amount and investors for a funding round, revenue figures for a solo-money story. Below 5 an item does not enter the digest. A press release scores 3 or less, a reprint loses 2, anything older than 14 days loses 1, a headline that misleads loses 3.
read the source