Patchwire

Moonshot AI · Released Jul 11, 2025

Kimi K2 0711

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for...

Open weights
Compare in showdown → OpenRouter ↗ All Moonshot AI models
Released
Jul 11 2025
Context window
131K 98K out
Input / 1M tokens
$0.57
Output / 1M tokens
$2.30

Where to run it

Providers

Only Novita serves Kimi K2 0711 on OpenRouter right now.

ProviderInputOutputContextMax outputUptimevs. best
NovitaFP8 $0.57 $2.30 131K98K100%24h Cheapest

Value

Price vs. performance

4 models score higher for the same price or lesson Aider Polyglot, e.g. DeepSeek V3.2.
11.1%36.7%62.4%88.0% $0.01$0.10$1.00$10.00$100.00 DeepSeek V3.2 (DeepSeek) — 74.2% · $0.28/1M in DeepSeek V3.2 Exp (DeepSeek) — 74.2% · $0.27/1M in GPT-5 (OpenAI) — 88.0% · $1.25/1M in gpt-oss-120b (OpenAI) — 41.8% · $0.15/1M in Codestral 2508 (Mistral) — 11.1% · $0.30/1M in Qwen3 235B A22B Instruct 2507 (Qwen) — 59.6% · $0.09/1M in o3 Pro (OpenAI) — 84.9% · $20.00/1M in R1 0528 (DeepSeek) — 71.4% · $0.50/1M in Qwen3 32B (Qwen) — 40.0% · $0.08/1M in Kimi K2 0905 (Moonshot AI) — 59.1% · $0.60/1M in Kimi K2 0711 (Moonshot AI) — 59.1% · $0.57/1M in Kimi K2 0711

Kimi K2 0711 Other Moonshot AI Other labs Value frontier · Aider Polyglot vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.

Benchmarks

1 benchmark score

BenchmarkScoreRankLeader
Aider PolyglotSoftware engineering · Hard Exercism problems in many languages 59.1% #7of 11 GPT-588.0%

Specs

Details

Model ID
moonshotai/kimi-k2
Developer
Moonshot AI
Release date
Jul 11, 2025 ·
Context window
131,072 tokens
Max output
98,304 tokens
API price
$0.57 / $2.30 per 1M input / output tokens
Providers
1
Input types
text
Reasoning
No
Tool calling
Yes
Open weights
Yes
Free tier
No

Release date, prices and providers from OpenRouter's catalog; buying directly from a provider can cost a little differently. Uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.