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 | 256k |
| 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 | ||
MATH-500Reasoning — accuracy on the 500-problem MATH subset | 91% | 95.4% |
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% | 53.2% |
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% | — |
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% | — |
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% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 77.2% | — |
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 | Mar 16 2026 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Mistral Medium 3vsClaude Fable 5.2Nemotron 3 Nano 4BvsClaude Fable 5.2Mistral Medium 3vsGPT-6.2 AstraNemotron 3 Nano 4BvsGPT-6.2 AstraMistral Medium 3vsGemini 4 NeonNemotron 3 Nano 4BvsGemini 4 NeonMistral Medium 3vsMuse EmberNemotron 3 Nano 4BvsMuse EmberMistral Medium 3vsGrok 4.7Nemotron 3 Nano 4BvsGrok 4.7Mistral Medium 3vsDeepSeek-V4.1-Flash-1017Nemotron 3 Nano 4BvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Mistral Medium 3 and Nemotron 3 Nano 4B are evenly matched across the 2 benchmarks they both report (MATH-500, GPQA Diamond). Mistral Medium 3 shipped 313 days before Nemotron 3 Nano 4B, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (Mistral Medium 3) vs 256k (Nemotron 3 Nano 4B). Mistral Medium 3 is closed, while Nemotron 3 Nano 4B is open weight.
On MATH-500, Nemotron 3 Nano 4B leads at 95.4% vs Mistral Medium 3 at 91%. On GPQA Diamond, Mistral Medium 3 leads at 57.1% vs Nemotron 3 Nano 4B at 53.2%.