modelgrep
M

MoonshotAI: Kimi K2.5

moonshotai/kimi-k2.5

53rd smartest of 178ReasoningToolsJSONVision
Use via OpenRouter ↗
Intelligence
37.3
53rd of 178
Design Elo
1293
Website
Speed
81
91st fastest
Latency
218ms
first token
Input price
$0.375
143rd cheapest
Context
262K

How it compares

Smarter than70%
of all ranked models
Faster than69%
of all ranked models
Cheaper than52%
of all ranked models

Overview

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...

Benchmarks

independent · via OpenRouter
Artificial Analysis68th percentile
Intelligence Index
37.3
Coding Index
25.8
Agentic Index
52.8
GPQA Diamond
79%
Humanity's Last Exam
12%
SciCode
40%
Tau²-Bench (agentic)
81%
Design Arena · Elo24,225 tournaments
Website
1293
3D
1292
UI Component
1291
codecategories
1289
Game Dev
1277
Data Viz
1272
godotgamedev
1257
asciiart
1227
svg
1214
webapps
1209
mobileapps
1195
fullstack
1193
androidnative
1138

Providers & pricing (12)

ProviderIn $/MOut $/MUptime
DigitalOcean$0.375$2.0299.7%
ModelRunfp4$0.400$1.9098.4%
Chutesint4$0.440$2.0084.9%
DeepInfrafp4$0.450$2.2599.9%
SiliconFlowint4$0.450$2.2599.8%
AtlasCloudint4$0.490$2.5099.7%
StreamLake$0.540$2.7099.7%
Venice$0.560$3.50
Novita$0.570$2.8599.9%
Phala$0.600$3.00
Moonshot AIint4$0.600$3.00100%
Fireworks$0.600$3.000%

Specifications

Context window262K
Max output
Knowledge cutoff
Input modalitiestext, image
Output modalitiestext
Prompt caching
Cache read price
ModeratedNo

Kimi K2.5 FAQ

How much does Kimi K2.5 cost?

Kimi K2.5 costs $0.375 per million input tokens and $2.02 per million output tokens via OpenRouter, making it 143rd cheapest of 298 paid models.

How smart is Kimi K2.5?

Kimi K2.5 scores 37.3 on the Artificial Analysis Intelligence Index, ranking 53rd of 178 benchmarked models, with a GPQA Diamond score of 79%.

How fast is Kimi K2.5?

Kimi K2.5 generates around 81 tokens per second with 218ms time-to-first-token (p50), the 91st fastest tracked model.

What is Kimi K2.5's context window?

Kimi K2.5 supports a 262K-token context window. It accepts text, image input.

Compare head-to-head