Compare AI models
| 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 | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $0.60 | $1.75 |
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. | $3.00 | $14.00 |
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. | $0.10 | $0.175 |
Cheapest inputLowest input rate across every provider on OpenRouter, the lab included. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.50SiliconFlow | $1.75Azure |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $2.50SiliconFlow | $14.00Azure |
| $225SiliconFlow · int4 | $1,050OpenAI list price | |
| Benchmarks | ||
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. | 52% | 24% |
SWE-Bench VerifiedCoding — Real 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% | 85% |
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. | 16% | 68% |
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. | 63.3% | 64.7% |
LiveCodeBenchCompetitive coding — Coding 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% | — |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 60.6% | — |
Humanity's Last Exam · with toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | 30.1% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 87.6% | — |
threejsevalCommunity preference (Three.js) — Every model gets the same prompt — "the Eiffel Tower", "a glass fishbowl", "a robot arm picking toys into a box" — and builds a 3D scene in Three.js. Real people then see two scenes side by side, names hidden, and vote for the one they prefer. The votes become a chess-style Elo rating on threejseval.com, averaged across all the prompts. It measures whether the scene looks and moves right to a human eye, not whether the code passes a test. Higher is better. | — | 1171 |
| Overview | ||
| Company | Moonshot AI | OpenAI |
| Release date | Jan 27 2026 | Feb 5 2026 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
Kimi K2.5vsClaude Haiku 5.5GPT-5.3-CodexvsClaude Haiku 5.5Kimi K2.5vsGemini 4 ArgonGPT-5.3-CodexvsGemini 4 ArgonKimi K2.5vsMuse WonderGPT-5.3-CodexvsMuse WonderKimi K2.5vsGrok 4.7GPT-5.3-CodexvsGrok 4.7Kimi K2.5vsDeepSeek-V4.1-FlashGPT-5.3-CodexvsDeepSeek-V4.1-FlashKimi K2.5vsMistral Large 4GPT-5.3-CodexvsMistral Large 4Frequently asked questions
GPT-5.3-Codex leads Kimi K2.5 on 3 of the 4 benchmarks they both report (BullshitBench v2, SWE-Bench Verified, Next.js Evals, OSWorld-Verified). Kimi K2.5 is cheaper on both input and output: $0.60 vs $1.75 per million input tokens, and $3.00 vs $14.00 per million output tokens. Kimi K2.5 shipped 9 days before GPT-5.3-Codex, so benchmark comparisons should account for the intervening progress.
Kimi K2.5 is open weight, while GPT-5.3-Codex is closed.
On BullshitBench v2, Kimi K2.5 leads at 52% vs GPT-5.3-Codex at 24%. On SWE-Bench Verified, GPT-5.3-Codex leads at 85% vs Kimi K2.5 at 76.8%. On Next.js Evals, GPT-5.3-Codex leads at 68% vs Kimi K2.5 at 16%. On OSWorld-Verified, GPT-5.3-Codex leads at 64.7% vs Kimi K2.5 at 63.3%.