Landmark Foundation Models · 1 min read
Claude 3 (Opus, Sonnet, Haiku): Anthropic surpasses GPT-4
In one sentence Anthropic ships the Claude 3 family in three sizes. Opus, the flagship, beats GPT-4 on MMLU, HumanEval, MATH. Native multimodal vision. For the first time GPT-4 is no longer the outright leader.
Reading level
Anthropic ships Claude 3 in three versions: Opus (top, smart, expensive), Sonnet (mid, balanced), Haiku (small, fast, cheap). All three see images.
The headline: Opus, on nearly every major benchmark, beats GPT-4. It's the first time in a year of GPT-4 dominance that another model takes the top spot.
For developers, it's the moment "default = GPT-4" stops being an obvious choice. The real frontier-comparison era begins.
Companies
Anthropic
Tools
Claude 3 Opus, Claude 3 Sonnet, Claude 3 Haiku
Tags
AnthropicClaude 3OpusSonnetHaikuVision
Sources