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. | 236B | 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. | 128k | 256k |
| Benchmarks | ||
MATH-500Reasoning — accuracy on the 500-problem MATH subset | 82.8% | 95.4% |
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. | 34.38% | — |
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% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | — | 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. | — | 53.2% |
| Overview | ||
| Company | DeepSeek | NVIDIA |
| Release date | Dec 10 2024 | Mar 16 2026 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek V2.5 (Revised)vsClaude Fable 5.2Nemotron 3 Nano 4BvsClaude Fable 5.2DeepSeek V2.5 (Revised)vsGPT-6.2 AstraNemotron 3 Nano 4BvsGPT-6.2 AstraDeepSeek V2.5 (Revised)vsGemini 4 NeonNemotron 3 Nano 4BvsGemini 4 NeonDeepSeek V2.5 (Revised)vsMuse EmberNemotron 3 Nano 4BvsMuse EmberDeepSeek V2.5 (Revised)vsGrok 4.7Nemotron 3 Nano 4BvsGrok 4.7DeepSeek V2.5 (Revised)vsMistral Large 4Nemotron 3 Nano 4BvsMistral Large 4Frequently asked questions
Nemotron 3 Nano 4B leads DeepSeek V2.5 (Revised) on 1 of the 1 benchmark they both report (MATH-500). DeepSeek V2.5 (Revised) shipped 461 days before Nemotron 3 Nano 4B, so benchmark comparisons should account for the intervening progress.
DeepSeek V2.5 (Revised) has 236B parameters, while Nemotron 3 Nano 4B has 4B. Context windows are 128k (DeepSeek V2.5 (Revised)) vs 256k (Nemotron 3 Nano 4B).
On MATH-500, Nemotron 3 Nano 4B leads at 95.4% vs DeepSeek V2.5 (Revised) at 82.8%.