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 | 109B |
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 | 10M |
| 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. | — | $0.08DeepInfra |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | — | $0.24DeepInfra |
| — | $35.00DeepInfra · fp8 | |
| 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% | 32.8% |
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% | 57.2% |
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | — | 20% |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | 30.6% | — |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 96% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 74.3% |
MGSMMultilingual math — solving grade-school word problems translated into multiple languages | — | 90.6% |
MathVistaMultimodal reasoning — mathematical reasoning about visual information | — | 70.7% |
MMMU-Pro · standard and vision meanMultimodal reasoning — average across the standard and vision sets | — | 52.2% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | — | 69.4% |
| Overview | ||
| Company | DeepSeek | Meta |
| Release date | Dec 1 2025 | Apr 5 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek-V3.2-SpecialevsClaude Fable 5.2LLaMA 4 ScoutvsClaude Fable 5.2DeepSeek-V3.2-SpecialevsGPT-6.2 AstraLLaMA 4 ScoutvsGPT-6.2 AstraDeepSeek-V3.2-SpecialevsGemini 4 HydrogenLLaMA 4 ScoutvsGemini 4 HydrogenDeepSeek-V3.2-SpecialevsGrok 4.7LLaMA 4 ScoutvsGrok 4.7DeepSeek-V3.2-SpecialevsMistral Large 4LLaMA 4 ScoutvsMistral Large 4DeepSeek-V3.2-SpecialevsKimi K3.1 ProLLaMA 4 ScoutvsKimi K3.1 ProFrequently asked questions
DeepSeek-V3.2-Speciale leads LLaMA 4 Scout on 2 of the 2 benchmarks they both report (LiveCodeBench, GPQA Diamond). LLaMA 4 Scout shipped 240 days before DeepSeek-V3.2-Speciale, so benchmark comparisons should account for the intervening progress.
DeepSeek-V3.2-Speciale has 671B parameters, while LLaMA 4 Scout has 109B. Context windows are 128k (DeepSeek-V3.2-Speciale) vs 10M (LLaMA 4 Scout).
On LiveCodeBench, DeepSeek-V3.2-Speciale leads at 88.7% vs LLaMA 4 Scout at 32.8%. On GPQA Diamond, DeepSeek-V3.2-Speciale leads at 85.7% vs LLaMA 4 Scout at 57.2%.