Storia in 2 fonti

Kimi K2.7-Code cuts tokens 30%, but skips independent benchmarks

Kimi K2.7-Code claims 30% fewer thinking tokens and a drop-in API swap path, but independent benchmarks show kernel regressions and no DeepSWE submission.

Raccontata da

cryptobriefing.com

venturebeat.com

Confronto fonti

2 prospettive sulla stessa storia

AI · summaries

venturebeat.comStai leggendo6 h fa

Kimi K2.7-Code cuts tokens 30%, but skips independent benchmarks

Kimi K2.7-Code claims 30% fewer thinking tokens and a drop-in API swap path, but independent benchmarks show kernel regressions and no DeepSWE submission.

originale

cryptobriefing.com14 h fa

Kimi AI releases open-source K2.7 Code model with 1 trillion parameters on APIs and Hugging Face

Moonshot AI releases Kimi K2.7-Code, a 1 trillion parameter open-source coding model with 30% fewer reasoning tokens and double-digit benchmark gains over

Leggi questa versione → originale

Timeline cronologica

venerdì 12 giugno 2026·cryptobriefing.com
Kimi AI releases open-source K2.7 Code model with 1 trillion parameters on APIs and Hugging Face
Moonshot AI releases Kimi K2.7-Code, a 1 trillion parameter open-source coding model with 30% fewer reasoning tokens and double-digit benchmark gains over
venerdì 12 giugno 2026·venturebeat.com
Kimi K2.7-Code cuts tokens 30%, but skips independent benchmarks
Kimi K2.7-Code claims 30% fewer thinking tokens and a drop-in API swap path, but independent benchmarks show kernel regressions and no DeepSWE submission.