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 | 2M |
| 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. | 88.7% | 34.3% |
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% | 59.1% |
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 | — | 86.5% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 96% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 75.8% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | — | 65.9% |
| Overview | ||
| Company | DeepSeek | |
| Release date | Dec 1 2025 | Sep 24 2024 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
DeepSeek-V3.2-SpecialevsClaude Fable 5.2Gemini 1.5 Pro-002vsClaude Fable 5.2DeepSeek-V3.2-SpecialevsGPT-6.2 AstraGemini 1.5 Pro-002vsGPT-6.2 AstraDeepSeek-V3.2-SpecialevsMuse EmberGemini 1.5 Pro-002vsMuse EmberDeepSeek-V3.2-SpecialevsGrok 4.7Gemini 1.5 Pro-002vsGrok 4.7DeepSeek-V3.2-SpecialevsMistral Large 4Gemini 1.5 Pro-002vsMistral Large 4DeepSeek-V3.2-SpecialevsKimi K3.1 ProGemini 1.5 Pro-002vsKimi K3.1 ProFrequently asked questions
DeepSeek-V3.2-Speciale leads Gemini 1.5 Pro-002 on 2 of the 2 benchmarks they both report (LiveCodeBench, GPQA Diamond). Gemini 1.5 Pro-002 shipped 433 days before DeepSeek-V3.2-Speciale, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (DeepSeek-V3.2-Speciale) vs 2M (Gemini 1.5 Pro-002). DeepSeek-V3.2-Speciale is open weight, while Gemini 1.5 Pro-002 is closed.
On LiveCodeBench, DeepSeek-V3.2-Speciale leads at 88.7% vs Gemini 1.5 Pro-002 at 34.3%. On GPQA Diamond, DeepSeek-V3.2-Speciale leads at 85.7% vs Gemini 1.5 Pro-002 at 59.1%.