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. | 1.5B/7B/8B/14B/32B/70B | 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 |
| 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. | — | $0.57Novita |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | — | $2.30Novita |
| — | $229Novita · fp8 | |
| Benchmarks | ||
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 65.2% | 75.1% |
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. | — | 10% |
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. | — | 65.8% |
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. | 57.5% | — |
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. | — | 4.7% |
MATH-500Reasoning — accuracy on the 500-problem MATH subset | 94.5% | — |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | 70% | — |
| Overview | ||
| Company | DeepSeek | Moonshot AI |
| Release date | Jan 20 2025 | Jul 11 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek R1 DistillvsClaude Fable 5.2Kimi K2vsClaude Fable 5.2DeepSeek R1 DistillvsGPT-6.2 AstraKimi K2vsGPT-6.2 AstraDeepSeek R1 DistillvsGemini 4 HydrogenKimi K2vsGemini 4 HydrogenDeepSeek R1 DistillvsMuse EmberKimi K2vsMuse EmberDeepSeek R1 DistillvsGrok 4.7Kimi K2vsGrok 4.7DeepSeek R1 DistillvsMistral Large 4Kimi K2vsMistral Large 4Frequently asked questions
Kimi K2 leads DeepSeek R1 Distill on 1 of the 1 benchmark they both report (GPQA Diamond). DeepSeek R1 Distill shipped 172 days before Kimi K2, so benchmark comparisons should account for the intervening progress.
DeepSeek R1 Distill has 1.5B/7B/8B/14B/32B/70B parameters, while Kimi K2 has 1T.
On GPQA Diamond, Kimi K2 leads at 75.1% vs DeepSeek R1 Distill at 65.2%.