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. | 671B | — |
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 |
| Benchmarks | ||
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 85.7% | 42.5% |
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. | 88.7% | — |
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. | — | 83.7% |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | 30.6% | — |
MATHReasoning — accuracy on the original competition mathematics test set | — | 64.3% |
DROPReasoning — three-shot token F1 on questions requiring reasoning over paragraphs | — | 83.2 |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 96% | — |
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. | — | 84.7% |
MGSMMultilingual math — solving grade-school word problems translated into multiple languages | — | 87.1% |
| Overview | ||
| Company | DeepSeek | OpenAI |
| Release date | Dec 1 2025 | Nov 6 2023 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
DeepSeek-V3.2-SpecialevsClaude Fable 5.2GPT-4 TurbovsClaude Fable 5.2DeepSeek-V3.2-SpecialevsGemini 4 HydrogenGPT-4 TurbovsGemini 4 HydrogenDeepSeek-V3.2-SpecialevsMuse EmberGPT-4 TurbovsMuse EmberDeepSeek-V3.2-SpecialevsGrok 4.7GPT-4 TurbovsGrok 4.7DeepSeek-V3.2-SpecialevsMistral Large 4GPT-4 TurbovsMistral Large 4DeepSeek-V3.2-SpecialevsKimi K3.1 ProGPT-4 TurbovsKimi K3.1 ProFrequently asked questions
DeepSeek-V3.2-Speciale leads GPT-4 Turbo on 1 of the 1 benchmark they both report (GPQA Diamond). GPT-4 Turbo shipped 756 days before DeepSeek-V3.2-Speciale, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (DeepSeek-V3.2-Speciale) vs 128k (GPT-4 Turbo). DeepSeek-V3.2-Speciale is open weight, while GPT-4 Turbo is closed.
On GPQA Diamond, DeepSeek-V3.2-Speciale leads at 85.7% vs GPT-4 Turbo at 42.5%.