Moonshot AIKimi K2.5

Is Kimi K2.5 out yet?

Yes.

Available

About Kimi K2.5

Native multimodal agentic refresh of the K2 line.

How we confirmed it

Released via official access

January 27, 2026 at 12:00 AM UTC

Release timeline

    Waiting for Moonshot AI's next model?

    One click creates your free account (or signs you into an existing one) and turns on alerts for Moonshot AI releases. There's no password to set, and you can unsubscribe any time.

    Benchmarks

    Coding#44

    Kimi K2.5 (Reasoning) · via BenchLM.ai · captured Aug 16, 2026 · source

    54.28

    Reasoning#49

    Kimi K2.5 · via BenchLM.ai · captured Jul 17, 2026 · source

    51

    Knowledge#44

    Kimi K2.5 · via BenchLM.ai · captured Aug 16, 2026 · source

    57.4

    Math#5

    Kimi K2.5 · via BenchLM.ai · captured Aug 16, 2026 · source

    62.5

    MultimodalGrounded#48

    Kimi K2.5 · via BenchLM.ai · captured Jul 17, 2026 · source

    62.5

    Multilingual#8

    Kimi K2.5 · via BenchLM.ai · captured Aug 16, 2026 · source

    38.2

    InstructionFollowing#12

    Kimi K2.5 · via BenchLM.ai · captured Aug 16, 2026 · source

    91

    Agentic#43

    Kimi K2.5 · via BenchLM.ai · captured Jul 14, 2026 · source

    49.5

    Humanity's Last Exam

    kimi-k2.5 · via Epoch AI · captured Aug 16, 2026 · source

    24.4

    FrontierCode Diamond

    Kimi K2.5 · via Epoch AI · captured Jul 24, 2026 · source

    1

    SWE-Bench Pro#23self-reported

    moonshotai/Kimi-K2.5 · via Hugging Face · captured Aug 16, 2026 · source

    50.7

    Sourced metrics mirrored with attribution - not community sentiment.

    Data from BenchLM.ai.

    Data from Epoch AI, “AI Benchmarking Hub” (CC BY 4.0).

    Vibe rating

    How this model actually feels to use, rated by the people here.

    No vibe ratings yet. Be the first.

    Sign in to rate this model.

    Model Pulse

    How is this model behaving today? One check-in per person per day: the running record of “did they nerf it”. Resets at midnight UTC.

    No check-ins yet today. Be the first to log how this model is behaving.

    Sign in to check in on this model.Sign in

    Community consensus

    An AI summary of the public comments here. Not a sourced fact.

    Not enough public feedback yet to summarize. Add a comment to help build the picture.

    Compare Kimi K2.5 with another model →

    Also waiting on: Gemini 3.5 Pro, Claude Fable 5.1, Claude Mythos 5.1, GLM-6

    Share

    Comments

    Missing a model, or wishing OutYet did something it doesn't? Tell us what to build.