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 |
| 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. | $0.256DeepInfra | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.712DeepInfra | — |
| $103StreamLake | — | |
| 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. | 40.5% | 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. | 59.1% | 59.1% |
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 42% | — |
MATHReasoning — accuracy on the original competition mathematics test set | — | 86.5% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 75.8% |
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. | 88.5% | — |
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 26 2024 | Sep 24 2024 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
DeepSeek V3vsClaude Fable 5.2Gemini 1.5 Pro-002vsClaude Fable 5.2DeepSeek V3vsGPT-6.2 AstraGemini 1.5 Pro-002vsGPT-6.2 AstraDeepSeek V3vsMuse EmberGemini 1.5 Pro-002vsMuse EmberDeepSeek V3vsGrok 4.7Gemini 1.5 Pro-002vsGrok 4.7DeepSeek V3vsMistral Large 4Gemini 1.5 Pro-002vsMistral Large 4DeepSeek V3vsKimi K3.1 ProGemini 1.5 Pro-002vsKimi K3.1 ProFrequently asked questions
DeepSeek V3 leads Gemini 1.5 Pro-002 on 1 of the 2 benchmarks they both report (LiveCodeBench, GPQA Diamond). Gemini 1.5 Pro-002 shipped 93 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (DeepSeek V3) vs 2M (Gemini 1.5 Pro-002). DeepSeek V3 is open weight, while Gemini 1.5 Pro-002 is closed.
On LiveCodeBench, DeepSeek V3 leads at 40.5% vs Gemini 1.5 Pro-002 at 34.3%. On GPQA Diamond, both models score 59.1%.