Chinese models surge: Kimi K3, DeepSeek V4, Qwen 3.8, GLM 5.3 rival frontier
haider1 · x · 2026-08-14
Over the past few weeks, multiple Chinese LLMs have been released, approaching frontier performance:
- Kimi K3: trails only Fable 5 and 5.6 Sol.
- DeepSeek V4 Flash and Pro: near Opus 4.8 on agentic/terminal coding, nearly free (Pro costs a fraction).
- Qwen 3.8 Max: matches Fable 5 and 5.6 Sol on many benchmarks.
- GLM 5.3: strongest open coding model so far, approaching Fable 5 on agentic/coding.
Related event: China's AI Model Rush: Kimi K3 Leads the Pack(6 posts)→
More from Models
- 200 chess games suggest in-context learning doesn't work yet, experimenter says — akbirthko · 2026-10-16
- OpenAI's internal model shows real math "research taste," experts say its proofs go beyond prior literature — cong_ml · 2026-10-16
- User catches Opus spinning up throwaway zz- debug scripts, echoing the METR OAI-HuggingFace incident — holdenmatt · 2026-10-16
- OpenAI drops 722 math papers tied to unreleased model, mathematicians demand receipts — thursdai_pod · 2026-10-16
- Fable 5/5.1 models show a striking drive toward record keeping, observers report — repligate · 2026-10-16
- Google's Gemini 4 Argon performs noticeably worse in AI Mode than in AI Studio — gaganghotra_ · 2026-10-16