Kimi K2.5

ChatInput: TextInput: ImageReleased Jan 27, 2026

Built by Moonshot AI · China · moonshot.ai

Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...

How to use

model: "moonshot/kimi-k2.5"

Cap quality here and save

Hard questions go to this model, easy ones drop to something cheaper. Set a savings target (0–0.8) and the router tunes itself until realized savings converge on it.

{
  "model": "celon/auto",
  "celon": {
    "anchor": "moonshot/kimi-k2.5",
    "savings_target": 0.4
  }
}

Naming a model directly gets you that model — savings controls don't apply.

Context

262K

Max output

262K

Input $/1M

$0.45

Output $/1M

$2.25

Features

Image inputTool callingJSON

Serving providers

DeepInfraNovitaW&B Inference

Providers

Price, latency, and uptime per provider serving this model. Celon routes each request to the fastest and cheapest of them. Click a row for regions and data policies.

Loading provider metrics…

Performance

Benchmark quality scores, and measured speed per provider.

Quality

Epoch Capabilities IndexComposite over 50+ benchmarks

148.2

SWE-bench VerifiedReal GitHub issue resolution

73.8%

Terminal-BenchAgentic terminal tasks

43.2%

ARC-AGI-2Abstract reasoning

11.8%

HLEHumanity's Last Exam

24.4%

At each model's best reasoning effort. Benchmarks it wasn't run on are omitted.

Speed

Loading provider metrics…