Moonshot AI Kimi k2.6

3 appearances · meta-score 70.0/100 · last seen 2026-07-17 · scored under rubric v15

Pooled composite across 3 primary runs: 55.5 ± 17.4 (95% CI)

$77.11Total cost
128,253,181Tokens
48.9%Cache hit
2,549Requests
0.16%Failure rate
1.17Tools/response
1,663,293Tokens per $

Agent role scores

Per-agent-role scores for Moonshot AI Kimi k2.6
Agent role	Score / 10
Sprint-Review Agent	8.17
Code-Review Agent	7.79
Develop Agent	7.73
Test Agent	7.32
Orchestrator	7.28
Build Agent	7.13
Scout Agent	7.04
Sprint-Plan Agent	7.03
Code Agent	5.96
Pseudocode Agent	5.58
Platform Agent	4.44

Run history (primary model)

Runs where Moonshot AI Kimi k2.6 was the primary model, oldest first
Run	SoW	Date	Composite
Kimi k2.6	circles	Jun 2026	53.1
Kimi k2.6	circles	Jul 2026	49.9
Kimi k2.6	circles	Jul 2026	63.3

Every run on Favur Evals is scored by the same deterministic engine, on the same Statements of Work. The benchmark is self-funded — no vendor sponsorship, credits, or grants.