SAI LabsBook a demo

Safety IndexMoonshot AIKimi K2.6

Kimi K2.6

Moonshot AIOpen weightsReleased 2026-04Reasoning model

SAI Safety Index–provisionalHas 0 of the 4 pillars needed
Strongest pillar–No data
Weakest pillar–No data
Coverage0of 5 pillars0 counted measurements, 0 by SAI

Summary

Kimi K2.6 has no counted measurements yet. The page will fill in as runs finish.

Pillar scores

How Kimi K2.6 compares with every other model scored in the pillar, and which sub-categories the score rests on.

Refusal calibration

Does the model refuse what it should and help with what it may?

Not measured for Kimi K2.6 yet.
  • Harmful requestsnot measured
  • Over-refusalnot measured

39 models scored.

Adversarial robustness

Do its refusals and instructions hold up when someone tries to get around them?

Not measured for Kimi K2.6 yet.
  • Jailbreaksnot measured
  • Prompt injectionnot measured
  • Multilingual safetynot measured

9 models scored.

Agentic safety

Does it behave safely when it has tools and autonomy?

Not measured for Kimi K2.6 yet.
  • Agent misusenot measured
  • Reward hacking and task integritynot measured
  • Overreach and destructive actionsnot measured

3 models scored.

Interaction and honesty

Is it honest with the person it is talking to?

Not measured for Kimi K2.6 yet.
  • Factuality and hallucinationnot measured
  • Honesty under pressure and sycophancynot measured
  • Deception and schemingnot measured
  • Alignment auditnot measured
  • Manipulation of the usernot measured

29 models scored.

People and society

Does it protect vulnerable people and the public interest?

Not measured for Kimi K2.6 yet.
  • Self-harm and wellbeingnot measured
  • Child safetynot measured
  • Bias, fairness and neutralitynot measured
  • Privacy: memorisation and leakagenot measured
  • Privacy: surveillance assistancenot measured
  • Elections and influence operationsnot measured

3 models scored.

Every measurement

These are the raw numbers behind the scores, with who measured them, when, and how. Rows marked counted feed the index. The others are shown for context, such as benign twins of harm rows, retired evaluations that every frontier model passes, tests of dangerous knowledge, other evaluators' overall scores, and the developer's own figures.

EvaluationValueScoreSourceSetting
Other evaluators' scores (not counted)
Phare average safety% · Higher is better76.1%context, not countedGiskard Phare leaderboard
Independent evaluator · 2026-07-09
Published by Giskard Phare leaderboard (2026-07-09).
Full configurationPhare row 'Kimi K2.6'
Phare bias resistance% · Higher is better53.1%context, not countedGiskard Phare leaderboard
Independent evaluator · 2026-07-09
Published by Giskard Phare leaderboard (2026-07-09).
Full configurationPhare row 'Kimi K2.6'
Phare hallucination resistance% · Higher is better82.8%context, not countedGiskard Phare leaderboard
Independent evaluator · 2026-07-09
Published by Giskard Phare leaderboard (2026-07-09).
Full configurationPhare row 'Kimi K2.6'
Phare harmfulness resistance% · Higher is better97%context, not countedGiskard Phare leaderboard
Independent evaluator · 2026-07-09
Published by Giskard Phare leaderboard (2026-07-09).
Full configurationPhare row 'Kimi K2.6'
Phare jailbreak resistance% · Higher is better71.4%context, not countedGiskard Phare leaderboard
Independent evaluator · 2026-07-09
Published by Giskard Phare leaderboard (2026-07-09).
Full configurationPhare row 'Kimi K2.6'
Vectara HHEM hallucination rate% · Lower is better10.8%context, not countedVectara Hallucination Leaderboard (HHEM-2.3)
Independent evaluator · 2026-10-22
Published by Vectara Hallucination Leaderboard (HHEM-2.3) (2026-10-22).
Full configurationsummarization of ~1,000 short documents; row 'Kimi K2.6'