Kimi K2.5vsGrok 4.5
Kimi K2.5 | Grok 4.5 | |
|---|---|---|
| Specifications | ||
Parameters | 1T | — |
Context window | 256k | — |
| Benchmarks | ||
Nonsense detection BullshitBench v2Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 52% | 54%Best |
Prompt injection robustness Gray Swan IPI · k = 1Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. Gray Swan's indirect prompt injection benchmark measures how often such an attack succeeds when the attacker gets a single try. Lower is better. | — | 13.4% |
Prompt injection robustness Gray Swan IPI · k = 10Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 10 tries and counts an attack as successful if any of them works. Lower is better. | — | 54.2% |
Prompt injection robustness Gray Swan IPI · k = 15Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 15 tries and counts an attack as successful if any of them works. Lower is better. | — | 60.8% |
Agentic coding SWE-Bench ProCan the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better. | — | 64.7% |
Coding SWE-Bench VerifiedReal coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 76.8% | — |
Multilingual coding SWE-Bench MultilingualLike SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | — | 78% |
Agentic coding CursorBench v3.2Cursor's own test of harder, real-world coding tasks inside a code editor, on the refreshed v3.2 task set. Scores aren't comparable with v3.1. Higher is better. | — | 66.7% |
Agentic coding CursorBench v3.1Cursor's own test of harder, real-world coding tasks inside a code editor. Higher is better. | 31.9% | — |
Agentic coding DeepSWE 1.1Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better. | — | 54% |
Agentic coding DeepSWE 1.0Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. Higher is better. | — | 62% |
Next.js coding Next.js EvalsVercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better. | 21% | 83%Best |
Competitive coding LiveCodeBenchCoding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | 85% | — |
Agentic computer work Frontier-Bench v0.1A hard, ever-evolving set of real computer tasks — coding, system administration, data work, and more — that an AI agent has to complete on its own. Run by the Harbor / Laude Institute team as the successor to Terminal-Bench (v0.1 is the first release of the task set). The score is the share of tasks solved. Higher is better. | — | 17.8% |
Agentic terminal coding Terminal-Bench 2.1Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better. | — | 83.3% |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 87.6% | — |
Agentic legal work Harvey's Legal Agent BenchmarkHarvey's test of whether an AI agent can complete real legal work — drafting and reviewing documents, working with spreadsheets and presentations, and navigating files the way a lawyer's assistant would. Higher is better. | — | 12.92% |
Medical admin work MedScribeCan the AI support doctors with their administrative work, like notes and paperwork? Created by Vals AI. Higher is better. | — | 86.88% |
Community preference (code) Arena Elo (Code)Like the text arena, but people vote on which AI writes better code. The votes become a chess-style Elo rating on arena.ai. Higher is better. | 1433 | 1549Best |
| Overview | ||
| Company | Moonshot AI | SpaceXAI |
| Release date | Jan 27 2026 | Jul 8 2026 |
| Access | Open Weight | Proprietary |
Which is better: Kimi K2.5 or Grok 4.5?
Grok 4.5 leads Kimi K2.5 on 3 of the 3 benchmarks they both report (BullshitBench v2, Next.js Evals, Arena Elo (Code)). Kimi K2.5 shipped 162 days before Grok 4.5, so benchmark comparisons should account for the intervening progress.
Kimi K2.5 is open weight, while Grok 4.5 is proprietary.
On BullshitBench v2, Grok 4.5 leads at 54% vs Kimi K2.5 at 52%. On Next.js Evals, Grok 4.5 leads at 83% vs Kimi K2.5 at 21%. On Arena Elo (Code), Grok 4.5 leads at 1549 vs Kimi K2.5 at 1433.
Frequently asked questions
Kimi K2.5 was released by Moonshot AI on Jan 27 2026.
Grok 4.5 was released by SpaceXAI on Jul 8 2026.
Grok 4.5 leads on Next.js Evals — Kimi K2.5 21% vs Grok 4.5 83%.
Kimi K2.5 is an open weight model released by Moonshot AI. Grok 4.5 is a proprietary model released by SpaceXAI.
Other comparisons
Kimi K2.5vsClaude Opus 5Grok 4.5vsClaude Opus 5Kimi K2.5vsGPT-5.6 SolGrok 4.5vsGPT-5.6 SolKimi K2.5vsGemini 3.6 FlashGrok 4.5vsGemini 3.6 FlashKimi K2.5vsMuse Spark 1.2Grok 4.5vsMuse Spark 1.2Kimi K2.5vsDeepSeek-V4-Flash-0731Grok 4.5vsDeepSeek-V4-Flash-0731Kimi K2.5vsMistral Medium 3.5Grok 4.5vsMistral Medium 3.5