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. | — | 49B |
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. | $0.44Mistral | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $2.20Mistral | — |
| $198Mistral | — | |
| Benchmarks | ||
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. | 30.3% | 73.58% |
MATH-500Reasoning — accuracy on the 500-problem MATH subset | 91% | 97.4% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 57.1% | 71.97% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 77.2% | 79.53% |
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. | 92.1% | — |
IFEvalAgentic tasks — following verifiable instructions in generated responses | 89.4% | — |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | — | 7.64% |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | — | 87.5% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | — | 82.71% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 66.1% | — |
| Overview | ||
| Company | Mistral | NVIDIA |
| Release date | May 7 2025 | Jul 25 2025 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Mistral Medium 3vsClaude Fable 5.2Llama Nemotron Super 49B v1.5vsClaude Fable 5.2Mistral Medium 3vsGPT-6.2 AstraLlama Nemotron Super 49B v1.5vsGPT-6.2 AstraMistral Medium 3vsGemini 4 NeonLlama Nemotron Super 49B v1.5vsGemini 4 NeonMistral Medium 3vsMuse EmberLlama Nemotron Super 49B v1.5vsMuse EmberMistral Medium 3vsGrok 4.7Llama Nemotron Super 49B v1.5vsGrok 4.7Mistral Medium 3vsDeepSeek-V4.1-Flash-1017Llama Nemotron Super 49B v1.5vsDeepSeek-V4.1-Flash-1017Frequently asked questions
Llama Nemotron Super 49B v1.5 leads Mistral Medium 3 on 4 of the 4 benchmarks they both report (LiveCodeBench, MATH-500, GPQA Diamond, MMLU-Pro). Mistral Medium 3 shipped 79 days before Llama Nemotron Super 49B v1.5, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (Mistral Medium 3) vs 128k (Llama Nemotron Super 49B v1.5). Mistral Medium 3 is closed, while Llama Nemotron Super 49B v1.5 is open weight.
On LiveCodeBench, Llama Nemotron Super 49B v1.5 leads at 73.58% vs Mistral Medium 3 at 30.3%. On MATH-500, Llama Nemotron Super 49B v1.5 leads at 97.4% vs Mistral Medium 3 at 91%. On GPQA Diamond, Llama Nemotron Super 49B v1.5 leads at 71.97% vs Mistral Medium 3 at 57.1%. On MMLU-Pro, Llama Nemotron Super 49B v1.5 leads at 79.53% vs Mistral Medium 3 at 77.2%.