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. | 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 | 200k |
| Benchmarks | ||
MATH-500Reasoning — accuracy on the 500-problem MATH subset | 96.2% | 97.5% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 55.1% | 77.6% |
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. | — | 8% |
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. | — | 97.3% |
IFEvalAgentic tasks — following verifiable instructions in generated responses | 75.5% | — |
DROPReasoning — three-shot token F1 on questions requiring reasoning over paragraphs | — | 77.7 |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 46.3% | — |
SimpleQAKnowledge — accuracy on short, fact-seeking questions in the original SimpleQA set | — | 20.2% |
MMLUGeneral knowledge — A 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better. | — | 90% |
MGSMMultilingual math — solving grade-school word problems translated into multiple languages | — | 93.7% |
| Overview | ||
| Company | NVIDIA | OpenAI |
| Release date | May 20 2025 | Apr 16 2025 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
Llama Nemotron Nano 4B v1.1vsClaude Fable 5.2o4-minivsClaude Fable 5.2Llama Nemotron Nano 4B v1.1vsGemini 4 Neono4-minivsGemini 4 NeonLlama Nemotron Nano 4B v1.1vsMuse Embero4-minivsMuse EmberLlama Nemotron Nano 4B v1.1vsGrok 4.7o4-minivsGrok 4.7Llama Nemotron Nano 4B v1.1vsDeepSeek-V4.1-Flash-1017o4-minivsDeepSeek-V4.1-Flash-1017Llama Nemotron Nano 4B v1.1vsMistral Large 4o4-minivsMistral Large 4Frequently asked questions
o4-mini leads Llama Nemotron Nano 4B v1.1 on 2 of the 2 benchmarks they both report (MATH-500, GPQA Diamond). o4-mini shipped 34 days before Llama Nemotron Nano 4B v1.1, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (Llama Nemotron Nano 4B v1.1) vs 200k (o4-mini). Llama Nemotron Nano 4B v1.1 is open weight, while o4-mini is closed.
On MATH-500, o4-mini leads at 97.5% vs Llama Nemotron Nano 4B v1.1 at 96.2%. On GPQA Diamond, o4-mini leads at 77.6% vs Llama Nemotron Nano 4B v1.1 at 55.1%.