Benchmarks spark debate: Anthropic and OpenAI claimed months ahead of Meta and Google

scaling01 · x · 2026-10-13

scaling01 shares a benchmark chart arguing Anthropic and OpenAI are multiple months ahead of Meta and Google. The chart shows Claude Opus 4.5 at 58, Claude Sonnet 4.5 and Muse Wonder at 56, with Claude 51, GPT-6 and Astra Gemini 4 at 53. The claim reflects an ongoing debate over the frontier lab capability gap.

Related event: Benchmark Chart Sparks Debate: Anthropic and OpenAI Lead Meta and Google by Months(2 posts)→

Original post →

More from Models

• Muse Code coding model impresses in early testing: faster than Claude Code — QuixiAI · 2026-10-13
• "Another benchmaxxed release in 14 hours": AI circle mocks benchmark hype — thesaraharminta · 2026-10-13
• New SOTA on image pointing gets Scale AI CEO's endorsement — alexandr_wang · 2026-10-13
• "I witnessed Grok creating unspeakable horrors in the group chat" — Kyrannio · 2026-10-13
• Asking ChatGPT to show if Meta Wonder is good yields an instant 'Yes. Very.' — rose_e_wang · 2026-10-13
• Leak: OpenAI prepping GPT-6.1 Astra, successor to GPT-6 Astra, internal id gpt-6.1-astra — lyraxana · 2026-10-13

Models channel →