MoonshotAI: Kimi K2.5, Other — route, pricing & source
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...
OpenRoutermediumLast checked: May 14, 2026
Input
$0.4000 / 1M
/ 1M tokens
Output
$1.90 / 1M
/ 1M tokens
Cached input
Not available
Not available
Cache write
Not available
Not available
Context and limits
Context window
262.1K
Max output
262,144
Input modalities
text, image
Output modalities
text
Weight access
Closed
Availability
Unknown
Capability flags
VisionToolsJSONReasoningLong context
From source metadata + verification fields — confirm provider docs before production.
Benchmarks
- LMArena Text Arenaoverall · official_benchmark_api1,449Independent
Rank: 41 | Votes: 27282 | 95% CI: 1445.0-1453.6 | Published: 2026-05-07 | Model license label in dataset: Modified MIT
- LMArena WebDev Arenaoverall · official_benchmark_api1,430Independent
Rank: 26 | Votes: 9307 | 95% CI: 1423.1-1437.3 | Published: 2026-05-07 | Model license label in dataset: Modified MIT
- LMArena Vision Arenaoverall · official_benchmark_api1,247Independent
Rank: 22 | Votes: 12824 | 95% CI: 1239.4-1253.9 | Published: 2026-05-07 | Model license label in dataset: Modified MIT
Access and license
Official API
No
Open weights
No
Self-hostable
No
License
Open weightsModified MITWeights are published under vendor-specific terms (Llama/Qwen‑style licenses). Usually you may download/run locally under those terms; it may differ from MIT/Apache wording.
Example costs
Scenario: 1K input + 300 output tokens / request
10,000 requests
$9.70
100,000 requests
$97.00
1,000,000 requests
$970.00
Cheaper alternatives
Lower-cost models for the same scenario
Source
Source
OpenRouter Models API
Confidence
medium
Route
OpenRouter
Last checked
May 14, 2026
Source notes
- Includes Hugging Face benchmark data only when the dataset is public, ungated, allowlisted, and matched by reviewed aliases.
