# Open weights caught the frontier. The moat is leaking.

> By 2026 free, downloadable models from China rivalled the closed frontier on many tasks.

*When you can download a near-top-tier model for free, 'exclusive access' stops being a business.*

By The InsidersFeed Desk · InsidersFeed
Canonical: https://insidersfeed.com/news/open-weights-caught-frontier-moat-leaking

> **Key:** **The take:** the closed labs' real product was never just the model — it was *exclusive access* to a capability you couldn't get elsewhere. In 2026, you increasingly can get it elsewhere, for free, to run on your own box. That's a leak in the moat, and it doesn't reseal.

Roll the tape: **DeepSeek V4** (April, MIT) back near the top; **Qwen** the most-downloaded family on earth; **Kimi** (K2.6 to K2.7-Code) on a two-month cadence; **MiniMax M3** claiming frontier coding; **GLM** a strong all-rounder. Most from Chinese labs. None of them has to *beat* GPT-5.5 — they just have to be good enough that self-hosting becomes the rational choice.

## What actually erodes

Pricing power. When a 'good enough' open model exists, the closed incumbent can't charge a monopoly rent for the median task — only for the genuine frontier edge. So the closed labs get pushed *upmarket*, into the hardest reasoning and longest agent runs, while the open tier eats the commodity middle. That's a worse business than 'we own all the inference'.

> **Note:** **Don't overclaim it:** the top of the frontier is still closed, lots of open benchmarks are vendor-marketing, and 'open-weight' sometimes means 'weights next week' (hi, MiniMax M3). The closed labs aren't dead — they're being forced to actually be ahead, not just exclusive.

The geopolitics is the kicker: the open floor is being set largely by Chinese labs, which means the world's cheapest capable AI increasingly ships from Hangzhou and Shanghai, not San Francisco. However the model race ends, the open-weight wave already changed who gets to build — and that's the part that won't reverse.

## FAQ

### Do open-weight models threaten OpenAI and Anthropic?
They threaten pricing power for everyday tasks, not the companies outright. As capable open models proliferate, closed labs get pushed upmarket toward the hardest reasoning and longest agentic work, where they still lead — but they can charge a premium for less of the market.

### Why are so many top open models from China?
Chinese labs (Alibaba/Qwen, DeepSeek, Moonshot, MiniMax, Z.ai) made aggressive open-weight releases a core strategy, building ecosystems and mindshare. That shifted the open-source centre of gravity, so much of the world's cheapest capable AI now comes from Chinese labs.