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. | — | 51B |
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. | 32k | 128k |
| Benchmarks | ||
ARC-ChallengeReasoning — multiple-choice science questions from the AI2 Reasoning Challenge; unrelated to ARC-AGI | 94.2% | 69.2% |
HellaSwagReasoning — commonsense completion of short scenarios | 89.2% | 85.58% |
WinoGrandeReasoning — resolving ambiguous pronouns in context | 86.7% | 84.53% |
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. | 81.2% | 80.2% |
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. | 81% | 91.43% |
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. | 45.1% | — |
| Overview | ||
| Company | Mistral | NVIDIA |
| Release date | Feb 26 2024 | Sep 23 2024 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Mistral LargevsClaude Fable 5.2Llama 3.1 Nemotron 51BvsClaude Fable 5.2Mistral LargevsGPT-6.2 AstraLlama 3.1 Nemotron 51BvsGPT-6.2 AstraMistral LargevsGemini 4 NeonLlama 3.1 Nemotron 51BvsGemini 4 NeonMistral LargevsMuse EmberLlama 3.1 Nemotron 51BvsMuse EmberMistral LargevsGrok 4.7Llama 3.1 Nemotron 51BvsGrok 4.7Mistral LargevsDeepSeek-V4.1-Flash-1017Llama 3.1 Nemotron 51BvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Mistral Large leads Llama 3.1 Nemotron 51B on 4 of the 5 benchmarks they both report. Mistral Large shipped 210 days before Llama 3.1 Nemotron 51B, so benchmark comparisons should account for the intervening progress.
Context windows are 32k (Mistral Large) vs 128k (Llama 3.1 Nemotron 51B). Mistral Large is closed, while Llama 3.1 Nemotron 51B is open weight.
On ARC-Challenge, Mistral Large leads at 94.2% vs Llama 3.1 Nemotron 51B at 69.2%. On HellaSwag, Mistral Large leads at 89.2% vs Llama 3.1 Nemotron 51B at 85.58%. On WinoGrande, Mistral Large leads at 86.7% vs Llama 3.1 Nemotron 51B at 84.53%. On MMLU, Mistral Large leads at 81.2% vs Llama 3.1 Nemotron 51B at 80.2%. On GSM8K, Llama 3.1 Nemotron 51B leads at 91.43% vs Mistral Large at 81%.