---
title: "Kimi K2 1T Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Kimi K2."
source: "https://rankedagi.com/models/kimi-k2-1t"
---

# Kimi K2

| Field | Value |
| --- | --- |
| Organization | Moonshot |
| License | Open Source |
| Size | 1T |
| Version | old |
| Released | 2025-07-11 |
| Context window | 131K |
| Input cost per million tokens | $0.6 |
| Output cost per million tokens | $2.5 |
| Cached input cost per million tokens | $0.15 |
| Model ID | kimi-k2-0711-preview |
| Last updated | 2026-04-23T16:27:06.714Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 48.3% | RankedAGI Coding Score |  |
| SWEBench Verified | coding | 71.6% | Agentic Coding | https://www.swebench.com/index.html |
| RankedAGI Agentic | Agents | 38.8% | RankedAGI Agentic Score |  |
| LiveCodeBench v6 | coding | 53.7% | LiveCodeBench v6 - Evaluation of Large Language Models for Code |  |
| LiveBench Coding 25.5 | coding | 71.8% | LiveBench Coding Score 2025-05-30 |  |
| Aider Polyglot | coding | 60.0% | Aider Polyglot Code Completion Benchmark |  |
| RankedAGI Reasoning | reasoning | 41.5% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 4.7% | Multidisciplinary Reasoning (no tools) |  |
| GPQA Diamond | reasoning | 75.1% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| LiveBench Reasoning 25.5 | reasoning | 63.0% | LiveBench Reasoning Score 2025-05-30 |  |
| AIME 2025 | math | 49.5% | AIME 2025 Competition Math |  |
| AIME 2024 | math | 69.6% | AIME 2024 Competition Math |  |
| LiveBench Math Score | math | 74.4% | LiveBench Math Score 25-05 |  |
| MMLU Pro | knowledge | 81.1% | Graduate-level visual problem-solving |  |
| MMLU | knowledge | 89.5% | Image Reasoning |  |
| LiveBench IF 25.5 | instruction-following | 82.5% | LiveBench Instruction Following Score 2025-05-30 |  |
| LiveBench Average 25.5 | general | 62.7% | LiveBench Average Score 2025-05-30 |  |
| Instruction Following Evaluation | instruction-following | 89.8% | Instruction Following Evaluation |  |
| LiveBench Agentic Coding 25.5 | coding | 20.0% | LiveBench Agentic Coding Score 2025-05-30 | https://livebench.ai/ |
| RankedAGI Math | math | 54.5% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 45.8% | Overall RankedAGI score |  |
| GDPval-AA Elo | Agents | 529 | Office Tasks (Artificial Analysis) |  |
| SvelteBench old | coding | 84.4% | SvelteBench - Benchmark for Svelte (Old v1) | https://khromov.github.io/svelte-bench/v1/v1-benchmark-results-merged.html |
| Code DesignArena | Design | 1098 |  | https://www.designarena.ai/leaderboard/code |

## Model Links

| Title | URL |
| --- | --- |
| Release | https://moonshotai.github.io/Kimi-K2/ |