Kimi K2.7 CodevsGrok 4.5
Kimi K2.7 Code | Grok 4.5 | |
|---|---|---|
| Specifications | ||
ParametersA rough measure of how big the model is. More parameters usually means more capable and more expensive to run, though it is a poor guide on its own — a smaller, newer model often beats a larger, older one. | 1T | — |
Context windowHow much text the model can hold in mind at once — your question, any documents you attach, the conversation so far, and its own reply. Go past it and the earliest part falls out of view. | 256k | — |
| API pricingUSD per 1M tokens · lower wins · base tier | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $0.95 | $2.00 |
Output priceWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one. | $4.00 | $6.00 |
Cached input priceA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again. | $0.19 | $0.30 |
| Benchmarks | ||
CursorBench v3.2Agentic coding — Cursor'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. | 49.7% | 66.7% |
DeepSWE 1.1Agentic coding — Artificial 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. | 31% | 54% |
Next.js EvalsNext.js coding — Vercel'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. | 75% | 83% |
Arena Elo (Code)Community preference (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. | 1469 | 1555 |
| BenchmarksPublished by one model only | ||
BullshitBench v2Nonsense detection — Given 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. | — | 54% |
Gray Swan IPI · k = 1Prompt injection robustness — Attackers 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% |
Gray Swan IPI · k = 10Prompt injection robustness — Attackers 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% |
Gray Swan IPI · k = 15Prompt injection robustness — Attackers 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% |
SWE-Bench ProAgentic coding — Can 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% |
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | — | 78% |
DeepSWE 1.0Agentic coding — Artificial 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% |
FrontierCode v1.1 (Extended) · extended splitAgentic coding — frontier-difficulty agentic coding tasks (v1.1, extended split) | — | 56.6% |
APEX-SWEExpert software engineering — expert-level software-engineering tasks (AI Productivity Index) | — | 53.6% |
Frontier-Bench v0.1Agentic computer work — A 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% |
Terminal-Bench 3.0Agentic terminal coding — command-line task completion (v3.0, much harder task set) | — | 15.7% |
Terminal-Bench 2.1Agentic terminal coding — Can 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% |
APEX-AgentsExpert agentic work — expert-level agentic work tasks (AI Productivity Index) | — | 47.1% |
Harvey's Legal Agent BenchmarkAgentic legal work — Harvey'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% |
MedScribeMedical admin work — Can the AI support doctors with their administrative work, like notes and paperwork? Created by Vals AI. Higher is better. | — | 86.88% |
AA Intelligence IndexOverall intelligence — Artificial Analysis composite intelligence index across evals | — | 56 |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | — | 1526 |
AA-BriefcaseKnowledge work — Artificial Analysis agentic office-work eval (Elo) | — | 1313 |
Arena Elo (Text)Community preference — Real people chat with two anonymous AIs side by side and vote for the answer they prefer. Votes become a chess-style Elo rating on arena.ai — it measures which AI people actually like, not test scores. Higher is better. | — | 1468 |
| Overview | ||
| Company | Moonshot AI | SpaceXAI |
| Release date | Jun 12 2026 | Jul 8 2026 |
| Access | Open Weight | Proprietary |
Frequently asked questions
Grok 4.5 leads Kimi K2.7 Code on 4 of the 4 benchmarks they both report (CursorBench v3.2, DeepSWE 1.1, Next.js Evals, Arena Elo (Code)). Kimi K2.7 Code is cheaper on both input and output: $0.95 vs $2.00 per million input tokens, and $4.00 vs $6.00 per million output tokens. Figures are base-tier rates. Kimi K2.7 Code shipped 26 days before Grok 4.5, so benchmark comparisons should account for the intervening progress.
Kimi K2.7 Code is open weight, while Grok 4.5 is proprietary.
On CursorBench v3.2, Grok 4.5 leads at 66.7% vs Kimi K2.7 Code at 49.7%. On DeepSWE 1.1, Grok 4.5 leads at 54% vs Kimi K2.7 Code at 31%. On Next.js Evals, Grok 4.5 leads at 83% vs Kimi K2.7 Code at 75%. On Arena Elo (Code), Grok 4.5 leads at 1555 vs Kimi K2.7 Code at 1469.
Kimi K2.7 Code was released by Moonshot AI on Jun 12 2026.
Grok 4.5 was released by SpaceXAI on Jul 8 2026.
Grok 4.5 leads on CursorBench v3.2 — Kimi K2.7 Code 49.7% vs Grok 4.5 66.7%.
Kimi K2.7 Code is cheaper on both input and output: $0.95 vs $2.00 per million input tokens, and $4.00 vs $6.00 per million output tokens. Figures are base-tier rates. Rates are pay-as-you-go API prices verified on August 18, 2026.
Kimi K2.7 Code is an open weight model released by Moonshot AI. Grok 4.5 is a proprietary model released by SpaceXAI.
Other comparisons
Kimi K2.7 CodevsClaude Opus 5Grok 4.5vsClaude Opus 5Kimi K2.7 CodevsGPT-5.6-CyberGrok 4.5vsGPT-5.6-CyberKimi K2.7 CodevsGemini 3.7 FlashGrok 4.5vsGemini 3.7 FlashKimi K2.7 CodevsMuse GlimmerGrok 4.5vsMuse GlimmerKimi K2.7 CodevsDeepSeek-V4-Pro-0813Grok 4.5vsDeepSeek-V4-Pro-0813Kimi K2.7 CodevsMistral Medium 3.5Grok 4.5vsMistral Medium 3.5