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. | 1.5B/7B/8B/14B/32B/70B | 24B |
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 |
| 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. | — | $0.06DeepInfra |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | — | $0.16DeepInfra |
| — | $25.00DeepInfra · fp8 | |
| Benchmarks | ||
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 65.2% | 46.13% |
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. | 57.5% | — |
MATHReasoning — accuracy on the original competition mathematics test set | — | 69.42% |
MATH-500Reasoning — accuracy on the 500-problem MATH subset | 94.5% | — |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | 70% | — |
SimpleQAKnowledge — accuracy on short, fact-seeking questions in the original SimpleQA set | — | 12.1% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 69.06% |
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. | — | 80.5% |
MathVistaMultimodal reasoning — mathematical reasoning about visual information | — | 67.09% |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | — | 49.1% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | — | 62.5% |
| Overview | ||
| Company | DeepSeek | Mistral |
| Release date | Jan 20 2025 | Jun 20 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek R1 DistillvsClaude Fable 5.2Mistral Small 3.2vsClaude Fable 5.2DeepSeek R1 DistillvsGPT-6.2 AstraMistral Small 3.2vsGPT-6.2 AstraDeepSeek R1 DistillvsGemini 4 HydrogenMistral Small 3.2vsGemini 4 HydrogenDeepSeek R1 DistillvsMuse EmberMistral Small 3.2vsMuse EmberDeepSeek R1 DistillvsGrok 4.7Mistral Small 3.2vsGrok 4.7DeepSeek R1 DistillvsKimi K3.1 ProMistral Small 3.2vsKimi K3.1 ProFrequently asked questions
DeepSeek R1 Distill leads Mistral Small 3.2 on 1 of the 1 benchmark they both report (GPQA Diamond). DeepSeek R1 Distill shipped 151 days before Mistral Small 3.2, so benchmark comparisons should account for the intervening progress.
DeepSeek R1 Distill has 1.5B/7B/8B/14B/32B/70B parameters, while Mistral Small 3.2 has 24B.
On GPQA Diamond, DeepSeek R1 Distill leads at 65.2% vs Mistral Small 3.2 at 46.13%.