2025-12 · GPT-5 mainline
Launch: Gemini 3 线
Gemini 3 line joins the SOTA cluster. Multimodal + long-context remain Google’s wedge. Era-relative scores.
Score method：Scores are era-relative (month top ≈95–100) for landscape review — not official historical Artificial Analysis replays. From 2026-08, the latest month aligns with the live AA-style snapshot.
Month landscape board
| # | Model | Vendor | Era score | Elo≈ | Pricing | Highlight |
|---|---|---|---|---|---|---|
| 🥇 | GPT-5 线 | OpenAI | 99 | 1360 | 分档 | 新主线 |
| 🥈 | Claude 4.x | Anthropic | 97 | 1350 | 高 | — |
| 🥉 | Gemini 2.5/3 预览NEW | 94 | 1330 | 中 | — | |
| 4 | Grok 3/4 线 | xAI | 91 | 1310 | 中 | 实时 |
| 5 | 国产旗舰集群 | CN | 90 | 1297 | 低 | Kimi/Qwen/DeepSeek |
Elo≈ is a LMSYS-style discussion-scale estimate for recap — not an official historical snapshot.
Launches this month
- Gemini 3 线Googleapi
多模态全家桶
Pricing changes
No separately tagged pricing events (prior month scale carries over).
Benchmark context
Benchmark context: public discussion of Arena Elo, AA Intelligence, and open suites (GPQA / SWE-bench, etc.). This page is a narrative archive, not a live probe.
Some historical narratives remain bilingual; model names stay in original form.
Related tools
Clavue CLI / imux IDE / clavue-2.1
Clavue · CLI 与 Agent 平台
CLI / GUI Agent 运行时、设备登录、会员额度与 OpenAI 兼容 API（api.clavue.com）。开发脚本、CI 与本地工具一条链路。
打开 Clavue ↗imux · 原生 macOS AI IDE
Clavue 平台的原生 IDE：多 Agent 分屏、Ghostty 级终端、Agent Chat、浏览器自动化与 Supervisor——不是又一个 Electron 壳。
了解 imux ↗Clavue 2.1 · 产品级大模型
旗舰模型 clavue-2.1（及 fast / pro / rev）：适合复杂推理与长程 Agent 循环。Chat、imux、API 共用会员额度。
查看模型 ↗