Kimi K2.7-Code cuts thinking tokens 30% — but practitioners say the benchmarks don't check out

VentureBeatJune 12, 2026tech

Moonshot AI released Kimi K2.7-Code this week, an open-source update to its K2 coding model family, claiming leaner reasoning and double-digit performance gains. K2.7-Code is built on the same trillion-parameter mixture-of-experts architecture as its p redecessor K2.6 , and drops in via an OpenAI-compatible API — which matters for teams already running K2.6 in production gateways. When K2.6 launched in April, it topped OpenRouter's weekly LLM leaderboard — a ranking based on actual API routing d

This article was published on VentureBeat (venturebeat.com). Read the full article on the original source:

Read full article on VentureBeat

#negative #Technology #ai

Kimi K2.7-Code cuts thinking tokens 30% — but practitioners say the benchmarks don't check out

More from VentureBeat

Google researchers introduce 'faithful uncertainty,' allowing LLMs to offer best guesses instead of hallucinations

NanoClaw and JFrog launch 'immune system' to block AI agents from downloading malicious code

PixelRAG beats text parsers on accuracy and cuts AI agent token costs 10x

Microsoft’s open-source SkillOpt automatically upgrades AI agent skills without touching model weights

Xiaomi's new open source, agentic AI coding harness MiMo Code beats Claude Code at ultra-long, 200+ step tasks

Context compression finally works in production: new research cuts LLM input 16x without the accuracy hit

What AI benchmarks miss about real-world performance

Google's DiffusionGemma generates 256 tokens in parallel and self-corrects as it goes