Kimi K3 / Kimi Code · Epoch Automation Reports · Mean work-output rubric score · October 2026 ↗
Kimi / MODEL EVIDENCE
Kimi K3 / Kimi Code
2 benchmarks · 3 observations · 2 sources
Scores at a glance
2 benchmarks · large dot: Kimi K3 / Kimi Code; small dots: closest models where reported
xdev-Bench-hard ↗33.3%
Kimi K3 / Kimi Code · xdev-Bench-hard · pass@3 · Kimi K3.1 Pro · internal xdev-bench-hard ↗
1 more reported result
15.9% · Kimi K3 / Kimi Code · xdev-Bench-hard · Average success rate · Kimi K3.1 Pro · internal xdev-bench-hard · Introducing Kimi K3.1 Pro ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
There are not enough shared scored settings to identify close comparisons reliably.
Reports featuring this model
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
Names used in source tables
Kimi K3 / Kimi Code