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. | 1T | 4B |
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. | 256k | 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. | $0.60Novita | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $2.50Novita | — |
| $245Novita · fp8 | — | |
| Benchmarks | ||
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 51% | 78.5% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 74.2% | 53.2% |
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. | 69.2% | — |
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. | 55.9% | — |
τ²-bench · retailAgentic tasks — task success in the retail domain of the second-generation customer-service evaluation | — | 39.8% |
τ²-bench · airlineAgentic tasks — task success in the airline domain of the second-generation customer-service evaluation | — | 33.3% |
τ²-bench · telecomAgentic tasks — task success in the telecom domain of the second-generation customer-service evaluation | — | 33% |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 7.4% | — |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | 7.9% | — |
MATH-500Reasoning — accuracy on the 500-problem MATH subset | — | 95.4% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 81.9% | — |
| Overview | ||
| Company | Moonshot AI | NVIDIA |
| Release date | Sep 5 2025 | Mar 16 2026 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Kimi K2 (0905)vsClaude Fable 5.2Nemotron 3 Nano 4BvsClaude Fable 5.2Kimi K2 (0905)vsGPT-6.2 AstraNemotron 3 Nano 4BvsGPT-6.2 AstraKimi K2 (0905)vsGemini 4 NeonNemotron 3 Nano 4BvsGemini 4 NeonKimi K2 (0905)vsMuse EmberNemotron 3 Nano 4BvsMuse EmberKimi K2 (0905)vsGrok 4.7Nemotron 3 Nano 4BvsGrok 4.7Kimi K2 (0905)vsDeepSeek-V4.1-Flash-1017Nemotron 3 Nano 4BvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Kimi K2 (0905) and Nemotron 3 Nano 4B are evenly matched across the 2 benchmarks they both report (AIME 2025, GPQA Diamond). Kimi K2 (0905) shipped 192 days before Nemotron 3 Nano 4B, so benchmark comparisons should account for the intervening progress.
Kimi K2 (0905) has 1T parameters, while Nemotron 3 Nano 4B has 4B. Context windows are 256k (Kimi K2 (0905)) vs 256k (Nemotron 3 Nano 4B).
On AIME 2025, Nemotron 3 Nano 4B leads at 78.5% vs Kimi K2 (0905) at 51%. On GPQA Diamond, Kimi K2 (0905) leads at 74.2% vs Nemotron 3 Nano 4B at 53.2%.