Mapping the Largest LLMs: US vs China (November 2026)
A map of the largest language models from US and Chinese labs, ranked by estimated parameter count, in one table.
Last updated
Where the biggest models sit today, by lab and by country.
The map below lays out the largest LLMs from US and Chinese labs by estimated total parameters. TBA means announced or unreleased.
Source: Alan D. Thompson, lifearchitect.ai/models-table. Parameter counts are estimates, not lab disclosures.
| Model | Lab | Country | Est. size | Date |
|---|---|---|---|---|
| 10T | ByteDance | CN | 10T | TBA |
| Fable 5 (Mythos) | Anthropic | US | n/a | Jun/2026 |
| GPT-6.x | OpenAI | US | n/a | TBA |
| Claude Opus 5.x | Anthropic | US | ~5T | Jul/2026 |
| GPT-5.x Sol | OpenAI | US | ~5T | Jun/2026 |
| Gemini 3.x Pro | US | ~3T | Feb/2026 | |
| Kimi K3 | Moonshot AI | CN | 2.8T | Jul/2026 |
| MiniMax-M3-Pro | MiniMax | CN | 2.7T | TBA |
| ERNIE 5.x | Baidu | CN | 2.4T | Nov/2025 |
| Qwen 3.8 Max | Alibaba | CN | 2.4T | Jul/2026 |
| DeepSeek-V4-Pro | DeepSeek AI | CN | 1.6T | Apr/2026 |
| LongCat 2.0 | Meituan | CN | 1.6T | Jun/2026 |
| Grok 4.x | xAI | US | 1.5T | Jun/2026 |
| Yuan3.0-Ultra | YuanLabAI | CN | 1.5T | Mar/2026 |
| AFM 3 Cloud Pro | Apple | US | ~1.2T | Jun/2026 |
| MAI-Thinking-1 | Microsoft | US | 1T | Jun/2026 |
| Genesis-Science-1 | US DOE / Arcee AI | US | 1T | TBA |
| Mimo V2.x Pro | Xiaomi | CN | 1T | Apr/2026 |
| KAT-Coder-Pro | Kuaishou | CN | 1T | Jul/2026 |
| Intern-S1-Pro | Shanghai AI Lab | CN | 1T | Feb/2026 |
| Ring 2.x | Inclusion AI | CN | 1T | May/2026 |
| Inkling | Thinking Machines | US | 975B | Jul/2026 |
| Muse (Watermelon) | MSL (Meta) | US | n/a | TBA |