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. | 12B | 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. | 128k | 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. | $0.0166Io Net | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.0216Io Net | — |
| $4.40Io Net · fp16 | — | |
| Benchmarks | ||
HellaSwagReasoning — commonsense completion of short scenarios | 83.5% | 85.58% |
WinoGrandeReasoning — resolving ambiguous pronouns in context | 76.8% | 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. | 68% | 80.2% |
ARC-ChallengeReasoning — multiple-choice science questions from the AI2 Reasoning Challenge; unrelated to ARC-AGI | — | 69.2% |
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. | — | 91.43% |
| Overview | ||
| Company | Mistral | NVIDIA |
| Release date | Jul 18 2024 | Sep 23 2024 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Mistral NeMovsClaude Fable 5.2Llama 3.1 Nemotron 51BvsClaude Fable 5.2Mistral NeMovsGPT-6.2 AstraLlama 3.1 Nemotron 51BvsGPT-6.2 AstraMistral NeMovsGemini 4 NeonLlama 3.1 Nemotron 51BvsGemini 4 NeonMistral NeMovsMuse EmberLlama 3.1 Nemotron 51BvsMuse EmberMistral NeMovsGrok 4.7Llama 3.1 Nemotron 51BvsGrok 4.7Mistral NeMovsDeepSeek-V4.1-Flash-1017Llama 3.1 Nemotron 51BvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Llama 3.1 Nemotron 51B leads Mistral NeMo on 3 of the 3 benchmarks they both report (HellaSwag, WinoGrande, MMLU). Mistral NeMo shipped 67 days before Llama 3.1 Nemotron 51B, so benchmark comparisons should account for the intervening progress.
Mistral NeMo has 12B parameters, while Llama 3.1 Nemotron 51B has 51B. Context windows are 128k (Mistral NeMo) vs 128k (Llama 3.1 Nemotron 51B).
On HellaSwag, Llama 3.1 Nemotron 51B leads at 85.58% vs Mistral NeMo at 83.5%. On WinoGrande, Llama 3.1 Nemotron 51B leads at 84.53% vs Mistral NeMo at 76.8%. On MMLU, Llama 3.1 Nemotron 51B leads at 80.2% vs Mistral NeMo at 68%.