\ud83c\udfc1 AI竞赛时间线
追踪从2022年11月ChatGPT发布到2026年8月前沿模型的AI领导力演变。
当前领先者: Claude Mythos Preview
Claude Mythos Preview
Anthropic \u00b7 Score: 97.4
2022-11ChatGPT (GPT-3.5)
ChatGPT launch
OpenAI launches ChatGPT, starting the AI revolution.
查看Chatgpt 4o Latest 20250326详情 →2023-03GPT-4
GPT-4 release
GPT-4 sets a new standard for AI capability.
2023-07Claude 2
Claude 2 launch
Anthropic releases Claude 2 with 100K context window.
查看Claude Mythos Preview详情 →2024-03Claude 3 Opus
Claude 3 Opus
Claude 3 Opus briefly tops leaderboards.
查看Claude Mythos Preview详情 →2024-05GPT-4o
GPT-4o launch
GPT-4o brings multimodal capabilities.
2024-07Llama 3.1 405B
Llama 3.1 release
Meta releases the largest open-weight model.
查看Llama 3.1 Nemotron Ultra 253b V1详情 →2025-01DeepSeek R1
DeepSeek R1
DeepSeek releases R1, matching frontier models at a fraction of the cost.
查看Deepseek V4 Pro High Preview详情 →2025-06Gemini 2.5 Pro
Gemini 2.5 Pro
Google Gemini 2.5 Pro brings thinking capabilities.
查看Gemini 3.1 Pro Preview详情 →2025-12Claude Opus 4.5
Claude Opus 4.5
Claude Opus 4.5 pushes the frontier in coding.
查看Claude Mythos Preview详情 →2026-02DeepSeek V4 Pro
DeepSeek V4
DeepSeek V4 Pro matches frontier models at lower cost.
查看Deepseek V4 Pro High Preview详情 →2026-04Claude Opus 4.7
Claude Opus 4.7
Claude Opus 4.7 matches GPT-5.5 at the frontier.
查看Claude Mythos Preview详情 →2026-05Gemini 3.1 Pro
Gemini 3.1 Pro
Google Gemini 3.1 Pro joins the frontier.
查看Gemini 3.1 Pro Preview详情 →2026-07Claude Opus 5
Claude Opus 5
Claude Opus 5 reaches the top of multiple leaderboards.
查看Claude Mythos Preview详情 →AI竞赛的关键转折
- 2022\u20132023: OpenAI dominance with GPT-3.5 and GPT-4
- 2024: Anthropic and Google catch up; open-weight models emerge
- 2025: Reasoning models change the game; cost drops 10x
- 2026: Frontier is a 5-way race