Chinese AI start-up Moonshot has introduced the Kimi K3 large language model with one million context tokens. The model is designed to compete with Anthropic’s Opus 4.8 per competitive coding benchmarks reported in early tests. Benchmark leaks show Kimi K3 performance challenges Anthropic’s Claude Fable5 head-on. Industry watchers claim the release heightens the AI competition among Chinese and Western developers. Moonshot frames the model as a step toward narrowing the performance gap against leading global alternatives.