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. | 100k | 256k |
| API pricingUSD per 1M tokens · lower wins | ||
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. | — | $0.60Google |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | — | $2.50Google |
| — | $245Novita · bf16 | |
| These models have no shared benchmark scores. | ||
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. | — | 71.3% |
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. | 58.7% | — |
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. | 86.7% | — |
| Overview | ||
| Company | Anthropic | Moonshot AI |
| Release date | Aug 9 2023 | Nov 6 2025 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Claude Instant 1.2vsGPT-6.1 SolKimi K2 ThinkingvsGPT-6.1 SolClaude Instant 1.2vsGemini 4 ArgonKimi K2 ThinkingvsGemini 4 ArgonClaude Instant 1.2vsMuse WonderKimi K2 ThinkingvsMuse WonderClaude Instant 1.2vsGrok 4.7Kimi K2 ThinkingvsGrok 4.7Claude Instant 1.2vsDeepSeek-V4.1-FlashKimi K2 ThinkingvsDeepSeek-V4.1-FlashClaude Instant 1.2vsMistral Large 4Kimi K2 ThinkingvsMistral Large 4Frequently asked questions
Claude Instant 1.2 and Kimi K2 Thinking don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Claude Instant 1.2 shipped 820 days before Kimi K2 Thinking, so benchmark comparisons should account for the intervening progress.
Context windows are 100k (Claude Instant 1.2) vs 256k (Kimi K2 Thinking). Claude Instant 1.2 is closed, while Kimi K2 Thinking is open weight.
Direct benchmark comparisons are unavailable — Claude Instant 1.2 and Kimi K2 Thinking don't publish scores on any of the same benchmarks.