TOPIC: Model Distillation  [distillation]
Training a smaller or rival model on a stronger model outputs; at industrial scale and without permission it shades into illicit capability extraction.
Aliases: model distillation
Stories filed under this topic: 1

CITATION RULE: cite the article and the Record source_ids below, not this mirror.
========================================================================

## Open weights close the frontier coding gap
Edition: 2026-06-30  ·  Section: tech  ·  Epistemic: inference
Byline: Tinkerton · Policy Desk
URL: /editions/2026-06-30/articles/china-open-weights
Deck: As Washington defers a public GPT-5.6 release and limits access to its strongest systems, Z.ai’s GLM-5.2 is reported close to top US models on coding and agentic tasks, with a million-token context and tuning for Chinese chips. Open-weight share on a major router has climbed sharply at a fraction of closed-model prices.
Topics: china-ai, open-weight-models, ai-export-controls, frontier-models, distillation
Key numbers: GLM-5.2 context window (Reuters) = 1M tokens · Fraudulent accounts alleged (senators' letter, via CNBC) = ~25,000 · Claude exchanges alleged (22 Apr–5 Jun 2026) = ~28.8M · Trusted US orgs cleared for Mythos 5 (Reuters) = >100
Record source_ids: E1 | E2 | E3 | E4