MoonshotAI: Kimi K2 Thinking

moonshotai/kimi-k2-thinking

Kimi K2 Thinking is MoonshotAI's open-weight contender, ranking #76 of 235 on the TryAii Score. Its best showing is #10 on AIME-2025. API pricing starts at $0.6/$2.5 per million input/output tokens — a little under the going rate for its score class, with 4 independent hosts serving it. It's reasonably quick at roughly 46-54 tokens/sec.

Prompt $/1M

$0.600

Completion $/1M

$2.500

Context

262,144 tok

Best provider

67 tok/s · DeepInfra

Specs

Model id: moonshotai/kimi-k2-thinking
Display name: MoonshotAI: Kimi K2 Thinking
Modality: text->text
Tokenizer: Other
Instruct type: —
Context length: 262,144
Max completion: 100,352
Prompt $/1M: $0.600
Completion $/1M: $2.500
Free tier: no

Capabilities

Function callingno
JSON modeno
System promptno
Visionno
Moderatedno

Benchmark scores

Each axis is one benchmark, normalized to the best model in the field.

No benchmark data for this model.

Speed by provider

Provider	tok/s	TTFT	Total latency
DeepInfra	67	419 ms	576 ms
AtlasCloud	54	853 ms	1,431 ms
Googlefastest	51	761 ms	861 ms
Nebius	46	522 ms	679 ms
Novita	38	902 ms	1,070 ms
moonshotaifastest	—	—	—
Artificial Analysis	0	0 ms	—