China AI Hub China AI Hub

Models / Long-context Chinese AI models (1M tokens)

Long-context Chinese AI models (1M tokens)

Long-context models accept a million or more input tokens in a single request — enough for whole-repository code review, multi-document analysis and long agent traces. This hub lists models whose documented context window is 1,000,000 tokens or more.

14
models match this dimension
Counted automatically from the models database: context_window >= 1,000,000

Input context is cheap to advertise but the output ceiling and cache pricing decide real cost. Output limits range from 131,072 (GLM, Qwen) to 1,048,576 (Kimi K3) tokens — the difference between reading a long document and generating one.

Editorial position

Our read: a 1M-token window is table stakes and almost useless as a differentiator. What separates models is retrieval accuracy at depth, the price of a long call, and whether quality degrades before the window ends. Vendors quote the maximum; buyers should measure the effective depth on their own documents, because the last 200K tokens are usually where the claim quietly stops being true.

Pattern in the data

Distribution of this dimension across matching models
Models matching14
Open weights10 of 14
Providers represented7
Context window range977K – 1024K
Input price range /1M$0.15 – $26
Licence familiesMIT, proprietary, Apache-2.0, Kimi, MiniMax

Long context has become a threshold rather than a gradient: models cluster at 1M tokens or sit well below it, with little in between. That clustering is a pricing decision as much as a technical one — a 1M window changes serving cost, which is why the cheapest tiers often stop at 262K.

Matching models

GLM-5.2

zhipu-ai · deprecated

  • Context: 1,048,576 tokens
  • Open weight · MIT (pure open, no regional limits per official model card)

GLM-5.3

zhipu-ai · active

  • Context: 1,048,576 tokens
  • Open weight · Apache-2.0 (per GitHub repo metadata; README has no separate weights-license section - verify per-model HF cards before reuse)

GLM-5.3-Flash

zhipu-ai · active

  • Context: 1,048,576 tokens
  • Open weight · Apache-2.0 (per GitHub repo metadata; README has no separate weights-license section - verify per-model HF cards before reuse)

Kimi K3

moonshot-ai · active

  • Context: 1,048,576 tokens
  • Open weight · Kimi K3 License (permissive MIT-style, but Model-as-a-Service operators with >$20M aggregate revenue over any 12 months must sign a separate agreement; products with >100M MAU or >$20M monthly revenue must display 'Kimi K3' in the UI)

MiniMax-M3

minimax · active

  • Context: 1,048,576 tokens
  • Open weight · MiniMax Community License (custom): free for non-commercial use; commercial use requires prominent attribution 'Built with MiniMax M3' plus written authorization from MiniMax if yearly revenue exceeds US$20M (otherwise a one-time notice to [email protected])

Decision guides

Original research

Technology

This page is generated from the models database; the member list and count are derived from each model's frontmatter fields, not written by hand. See the full model database and the methodology.