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. | 100k | 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. | — | $0.256DeepInfra |
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 | ||
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. | 78.5% | 88.5% |
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% |
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% |
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. | 71.2% | — |
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% |
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. | 88% | — |
| Overview | ||
| Company | Anthropic | DeepSeek |
| Release date | Jul 11 2023 | Dec 26 2024 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Claude 2vsGPT-6.1 SolDeepSeek V3 ChatvsGPT-6.1 SolClaude 2vsGemini 4 ArgonDeepSeek V3 ChatvsGemini 4 ArgonClaude 2vsMuse Spark OpenDeepSeek V3 ChatvsMuse Spark OpenClaude 2vsGrok 4.7DeepSeek V3 ChatvsGrok 4.7Claude 2vsMistral Large 4DeepSeek V3 ChatvsMistral Large 4Claude 2vsKimi K3DeepSeek V3 ChatvsKimi K3Frequently asked questions
DeepSeek V3 Chat leads Claude 2 on 1 of the 1 benchmark they both report (MMLU). Claude 2 shipped 534 days before DeepSeek V3 Chat, so benchmark comparisons should account for the intervening progress.
Context windows are 100k (Claude 2) vs 128k (DeepSeek V3 Chat). Claude 2 is closed, while DeepSeek V3 Chat is open weight.
On MMLU, DeepSeek V3 Chat leads at 88.5% vs Claude 2 at 78.5%.