NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Kimi K2.6
moonshotai logo

Kimi K2.6

by moonshotai · Released Apr 2026

Open SourceMultimodal
43.7
avg score
Rank #182
Compare
Better than 42% of all models
Context
262K tokens (~131 books)
Input $/1M
$0.95
Output $/1M
$4.00
Type
multimodal
License
Open Source
Benchmarks
23 tested
Data as of
About

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...

Tested on 23 benchmarks · BenchGecko score 43.7. Top scores: Chatbot Arena Elo — Coding (1508.7%), Chatbot Arena Elo — Overall (1461.0%), OTIS Mock AIME 2024-2025 (96.1%).

Looking for similar performance at lower cost?
Gemini 2.0 Flash scores 44.4 (102% as good) at $0.10/1M input · 89% cheaper
Capabilities
coding
66.3
#24 globally
math
46.5
#129 globally
knowledge
48.3
#137 globally
general
22.5
#154 globally
speed
43.0
#51 globally
Benchmark Scores
Compare All
Tested on 23 benchmarks · Ranked across 6 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
codingCompare coding →
SWE-Bench verified

Real-world software engineering tasks from GitHub issues. Models must diagnose bugs and write patches that pass test suites. Human-verified subset of SWE-bench.

76.7·
WeirdML

Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.

55.9·
mathCompare math →
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

96.1·
FrontierMath-Tiers-1-3-v2-Private
57.2·
FrontierMath-2025-02-28-Private

Original research-level math problems created by professional mathematicians. Problems are unpublished and cannot be memorized.

39.0·
knowledgeCompare knowledge →
GPQA diamond

Graduate-level science questions written by PhD experts. Diamond subset contains questions where experts disagree, testing deep understanding.

87.7·
SimpleQA Verified

Simple factual questions with verified correct answers. Tests accuracy of basic knowledge retrieval. Low scores indicate hallucination.

34.9·
Chess Puzzles

Tactical chess puzzles testing pattern recognition and multi-move calculation. Measures strategic reasoning ability.

22.1·
Quick compare:Alibaba Qwen logovs Qwen 3.5 Plus (hosted 397B-A17B)OpenAI logovs GPT-3.5 Turbo (older v0613)Meta logovs Llama 3.1 405B
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Recently Happened
Kimi K2.6 pricing increased 119%
Oct 3, 2026
Kimi K2.6 pricing dropped 32%
Sep 29, 2026
Kimi K2.6 pricing increased 75%
Aug 23, 2026
Kimi K2.6 pricing increased 46%
Aug 17, 2026
Kimi K2.6 pricing dropped 32%
Aug 15, 2026
Kimi K2.6 pricing increased 20%
Jul 3, 2026
Similar Models
Alibaba Qwen logo
Qwen 3.5 Plus (hosted 397B-A17B)
Alibaba Qwen
43.7n/a
OpenAI logo
GPT-3.5 Turbo (older v0613)
OpenAI
43.8$1.00/1M
Meta logo
Llama 3.1 405B
Meta
43.8n/a
Links
Info
moonshotaiPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@moonshotai
Source Code
GitHubHugging Face
BenchGecko API
kimi-k2-6
Specifications
  • Typemultimodal
  • Context262K tokens (~131 books)
  • ReleasedApr 2026
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.006
Available On
moonshotai logomoonshotai$0.95
Categories
codingmathknowledgegeneralspeed
Learn More
context-windowtransformeropen weightstokens
Share & Export
Tweet
Related Models
Alibaba Qwen logoQwen 3.5 Plus (hosted 397B-A17B)
43.7
OpenAI logoGPT-3.5 Turbo (older v0613)
43.8
Meta logoLlama 3.1 405B
43.8
Google DeepMind logoGemini 2.0 Flash (Dec 2024)
44.1
Kimi K2.6 is an open-source multimodal AI model by moonshotai, released in April 2026. It has an average benchmark score of 43.7. Context window: 262K tokens.

Related Models

Qwen 3.5 Plus (hosted 397B-A17B) · Alibaba QwenGPT-3.5 Turbo (older v0613) · OpenAILlama 3.1 405B · MetaGemini 2.0 Flash (Dec 2024) · Google DeepMindQwen3.7 Plus · Alibaba Qwen

Benchmarks

Chatbot Arena Elo — CodingChatbot Arena Elo — OverallOTIS Mock AIME 2024-2025GPQA diamondDtbench

Related Pages

moonshotai · Providermoonshotai · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-10-05

  • Kimi K2.6 by moonshotai. BenchGecko score 43.7, rank 181 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.95 input · $4.00 output per 1M tokens (as of 2026-10-05).
  • Sold by 18 providers (as of 2026-10-05): Inceptron (int4) $0.47 in / $2.45 out · Chutes (int4) $0.50 in / $2.85 out · DigitalOcean $0.57 in / $2.40 out · Decart (fp4) $0.59 in / $2.47 out · StreamLake (fp8) $0.60 in / $2.52 out · and 13 more. Every provider
  • Gecko Tests: Who Are You B (Knows who made it) · Tokenizer Tax E (92% more tokens outside English).

How to cite · data as of 2026-10-05

Kimi K2.6 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/kimi-k2-6

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP