Moonshot’s Kimi K3 lands on Nebius as an OpenAI-compatible API with 1M context
rohanpaul_ai · x · 2026-07-28
Nebius Token Factory added Kimi K3 as an OpenAI-compatible API on day 0. The model is described as Moonshot AI’s open-weight frontier model with up to 1M tokens of context, a 2.8T-parameter MoE architecture, and 16 of 896 experts active per token.
The post also highlights its positioning for long-horizon coding, knowledge work, and reasoning. On the Intelligence Index, Kimi K3 is said to score 57, just 2 points behind GPT-5.6 Sol.
Related event: Moonshot's Kimi K3 Launches on Nebius with 1M Context(3 posts)→
More from Infra
- Engram's random reads don't suit SSDs; CPU-memory over NVLink could serve all 72 GPUs — bookwormengr · 2026-09-11
- 80% of the DIY LLM inference hype posters have already quit — it's brutally hard systems work — abhijithneil · 2026-09-11
- Hugging Face's Ultra Scale Playbook: a free book on training LLMs on GPU clusters — mdancho84 · 2026-09-11
- Is inference latency becoming the biggest bottleneck for production AI agents? — Euphoric_Sea632 · 2026-09-11
- LLM Serving Metrics Thread: Why TPOT and Uptime Make or Break User Experience — abhijithneil · 2026-09-11
- PlanetScale launches sharded Postgres: 768 servers acting as one, 1PB scale — dhruv2038 · 2026-09-11