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. | 16B | 12B |
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 |
| 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 | ||
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. | 55.7% | 68% |
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. | 57.3% | — |
MATHReasoning — accuracy on the original competition mathematics test set | 27.9% | — |
HellaSwagReasoning — commonsense completion of short scenarios | — | 83.5% |
WinoGrandeReasoning — resolving ambiguous pronouns in context | — | 76.8% |
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. | 72% | — |
| Overview | ||
| Company | DeepSeek | Mistral |
| Release date | May 16 2024 | Jul 18 2024 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek V2 LitevsClaude Fable 5.2Mistral NeMovsClaude Fable 5.2DeepSeek V2 LitevsGPT-6.2 AstraMistral NeMovsGPT-6.2 AstraDeepSeek V2 LitevsGemini 4 HydrogenMistral NeMovsGemini 4 HydrogenDeepSeek V2 LitevsMuse EmberMistral NeMovsMuse EmberDeepSeek V2 LitevsGrok 4.7Mistral NeMovsGrok 4.7DeepSeek V2 LitevsKimi K3.1 ProMistral NeMovsKimi K3.1 ProFrequently asked questions
Mistral NeMo leads DeepSeek V2 Lite on 1 of the 1 benchmark they both report (MMLU). DeepSeek V2 Lite shipped 63 days before Mistral NeMo, so benchmark comparisons should account for the intervening progress.
DeepSeek V2 Lite has 16B parameters, while Mistral NeMo has 12B. Context windows are 32k (DeepSeek V2 Lite) vs 128k (Mistral NeMo).
On MMLU, Mistral NeMo leads at 68% vs DeepSeek V2 Lite at 55.7%.