Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | 1B/4B/12B/27B | 1T |
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 128k | 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. | $0.04DeepInfra | $0.60Google |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.05Parasail | $2.50Google |
| $10.50Parasail · bf16 | $245Novita · bf16 | |
| These models have no shared benchmark scores. | ||
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. | 3% | — |
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% |
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. | 29.7% | — |
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. | 87.8% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 42.4% | — |
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. | 95.9% | — |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 64.9% | — |
| Overview | ||
| Company | Moonshot AI | |
| Release date | Mar 12 2025 | Nov 6 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Gemma 3vsClaude Haiku 5.5Kimi K2 ThinkingvsClaude Haiku 5.5Gemma 3vsGPT-6.1 SolKimi K2 ThinkingvsGPT-6.1 SolGemma 3vsMuse Spark OpenKimi K2 ThinkingvsMuse Spark OpenGemma 3vsGrok 4.7Kimi K2 ThinkingvsGrok 4.7Gemma 3vsDeepSeek-V4.1-Flash-1017Kimi K2 ThinkingvsDeepSeek-V4.1-Flash-1017Gemma 3vsMistral Large 4Kimi K2 ThinkingvsMistral Large 4Frequently asked questions
Gemma 3 and Kimi K2 Thinking don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Gemma 3 shipped 239 days before Kimi K2 Thinking, so benchmark comparisons should account for the intervening progress.
Gemma 3 has 1B/4B/12B/27B parameters, while Kimi K2 Thinking has 1T. Context windows are 128k (Gemma 3) vs 256k (Kimi K2 Thinking).
Direct benchmark comparisons are unavailable — Gemma 3 and Kimi K2 Thinking don't publish scores on any of the same benchmarks.