AIVentureBeatabout 8 hours ago
Kimi K2.7-Code cuts tokens 30%, benchmarks doubted
4 min read
Moonshot AI released Kimi K2.7-Code, an open-source update to its K2 coding model, claiming 30% fewer thinking tokens and double-digit performance gains. Practitioners question the benchmarks, noting K2.6 topped OpenRouter's leaderboard based on real developer usage, not self-reported scores.
Level
Hype check
Tap to vote and see what everyone thinks.
