Board/Moonshot/Kimi K2.6
K

Kimi K2.6

Moonshot AI·Apr 8, 2026N/A

Open weights

April's open-weight value shock: 80.2% SWE-bench Verified at just $0.60/M input made it the default 'good enough frontier' choice for cost-sensitive agent fleets.

Status
Generally available
License
Open weights · Custom
Context window
256K
Max output
64K
Input · $/M
$0.6
Output · $/M
$2.50
Cached input · $/M
$0.15
Blended · $/M
$1.07
Parameters
~1T MoE
Inputs
Text only

Benchmarks

Composite 80.2· mean of published scores

Reasoning & knowledge

  • GPQA Diamond

    no public score

  • MMLU-Pro

    no public score

  • Humanity's Last Exam

    no public score

  • ARC-AGI-2

    no public score

Coding

  • SWE-bench Verified

    80.2%

    dataset best 93.4
  • SWE-bench Pro

    no public score

  • LiveCodeBench

    no public score

Agents & computer use

  • Terminal-Bench 2

    no public score

  • OSWorld 2.0

    no public score

  • BrowseComp

    no public score

Math

  • AIME 2026

    no public score

Known strengths

  • 80.2% SWE-V at commodity pricing
  • Battle-tested agent stack

Primary sources

  • Kimi platform
Compare this model →

More from Moonshot

  • KKimi K3Jul 22, 2026 · 1M ctx
  • KKimi K2 ThinkingNov 6, 2025 · 256K ctx