GPT 6.1 Sol vs Claude Sonnet 5.5: 3.5x Cost Gap in Same-Task Test
11/03/2026 — 11/03, 14:19·1 sources (all)·1 reports (all)
Key takeaways
- A comparison test by Leiphone ran GPT 6.1 Sol and Claude Sonnet 5.5 through the same API provider on three tasks: a 3D escape room, policy r…
Background & analysis
On September 28, 2026, Anthropic released Claude Sonnet 5.5; the next day, OpenAI released GPT 6.1 Sol. According to Leiphone, both companies moved in a strikingly similar direction: compressing near-flagship model capability into a faster, cheaper tier built for agents and professional work. Sonnet 5.5 emphasizes completing well-scoped everyday tasks in fewer steps, while GPT 6.1 Sol targets complex coding, computer operation, and professional work.
Both models are listed on OpenRouter at identical API prices: $2 per 1M input tokens and $10 per 1M output tokens. Leiphone placed GPT 6.1 Sol and Claude Sonnet 5.5 on the same workbench, routing both through the same reseller, and asked them to complete three tasks back to back: building a 3D escape room, researching current policy, and producing a Nobel Prize in Chemistry explainer. GPT 6.1 Sol totaled $38.5, while Claude Sonnet 5.5 came to just $10.9 — a gap of roughly 3.5x, despite matching list prices.
On speed, each model won one of the three. For the 3D escape room, Sol took 33 minutes versus Sonnet's 74. On the policy investigation, the results flipped: Sonnet finished its report in 5 minutes, while Sol took close to an hour. Leiphone also compared end-to-end elapsed time across the tasks.
AI-generated from 1 reports · updated 14 hours ago
Related AI tools
- Heat index
- 62
- All sources
- 1
- All reports
- 1
- First seen
- 17 hours ago
Reports on this story headlines open the original
In a side-by-side test run through the same platform and API reseller, GPT 6.1 Sol and Claude Sonnet 5.5 tackled three tasks: building a 3D escape room, researching a recent policy, and writing a chemistry explainer. Both are priced identically at $2 per million input tokens and $10 per million output tokens, yet the runs cost $38.5 and $10.9 respectively, roughly a 3.5x difference. Speed split by task: Sol finished the 3D build in 33 minutes versus Sonnet's 74, while Sonnet completed the policy report in 5 minutes against nearly an hour for Sol.
雷峰网AI score 64
Other stories people are talking about
- Heat index 76Testing Claude Opus 5.5 across 12 front-end frameworks1 sources
- Heat index 69Logs measure how much Claude Code message threads speed up sessions1 sources
- Heat index 82Claude Skills guide: how they differ from MCP1 sources
- Heat index 81Gemini 4 Carbon released, benchmarks show slight lead over Claude1 sources
- Heat index 64User shares tips to reduce false positives in Claude abuse checks1 sources
How is heat calculated?About the methodHide
Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.
This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.
- Surge
- Discussion rising fast
- New
- First report within 6 hours
- Rising
- Still gathering discussion