Skip to content
llmwaves

Kimi K2.6

Moonshot AI · released Apr 20, 2026

Open weightsReasoningTool useStructured outputVision

Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and...

Specification

Context window
262K
Max output
262K
Knowledge cutoff
Not stated
Parameters
1058.6B
Licence
other
Serving providers
21
Moderated
No
Uptime
100.0%

Intelligence

45.1

93th percentile

Coding

61.8

Coding Index

Agentic

31.2

Agentic Index

Output speed

89 t/s

Median across providers

Latency

337ms

Time to first token

Cost per task

$0.23

Estimated

Benchmarks

Where the score comes from

The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.

Evaluation scores

Percentage correct · higher is better

  • τ²-bench (Telecom)
    95.9%
  • GPQA Diamond
    91.1%
  • AA-LCR (long context)
    76.7%
  • IFBench
    76.0%
  • SciCode
    53.5%
  • Terminal-Bench Hard
    43.9%
  • Humanity's Last Exam
    37.5%

An evaluation missing from this list was not run for this model — it is not a zero.

View as table
EvaluationScore
τ²-bench (Telecom)95.9%
GPQA Diamond91.1%
AA-LCR (long context)76.7%
IFBench76.0%
SciCode53.5%
Terminal-Bench Hard43.9%
Humanity's Last Exam37.5%

Against its peers

Intelligence Index · this model highlighted, nearest peers in grey

Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.

View as table
ModelIntelligence
Qwen3.7 Max46.7
GPT-5.3-Codex45.5
MiniMax M345.4
V4 Pro45.3
Kimi K2.645.1
GPT-5.243.3
Kimi K2.7 Code43.0
MiMo-V2.5-Pro42.9

Percentile among all indexed models

Intelligence93th
Coding80th
Agentic60th
Terminal work91th
Arena Elo93th

Pricing

What it costs to run

List prices per million tokens, plus what one representative task works out to.

List price

Input / 1M tokens
$0.589
Output / 1M tokens
$2.48
Cached input / 1M
$0.099
Blended 3:1
$1.06

One task, estimated

$0.23

Input tokens
50,000
Output tokens
80,000
Profile
Reasoning

Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).

Serving providers

Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.

Arena

Head-to-head generation quality

Elo from pairwise judgements, broken out by the kind of thing the model was asked to build.

Overall Elo
1,329
Win rate
60.3%
Strongest at
3D scenes
Tournaments
10,809

Elo by category

Dot position on a 1,175–1,350 scale · Elo has no meaningful zero

Agent categories (full-stack apps, mobile apps) are only scored for models tested in agent harnesses.

View as table
CategoryEloWin rate
3D scenes132960.3%
UI components129956.0%
Websites129755.5%
Game development129056.0%
Data visualisation128656.1%
SVG122551.5%
Mobile apps122154.9%
Full-stack apps119854.8%