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. | — | 236B |
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 |
| Benchmarks | ||
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. | 58.7% | 87.2% |
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. | — | 19% |
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. | — | 39.7% |
Okapi Multilingual MMLUMultilingual knowledge — five-shot accuracy on the Okapi translations of MMLU used in the Claude 3 model card | 61.2% | — |
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. | 86.7% | — |
| Overview | ||
| Company | Anthropic | DeepSeek |
| Release date | Aug 9 2023 | Jul 24 2024 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Claude Instant 1.2vsGPT-6.2 AstraDeepSeek Coder V2-0724vsGPT-6.2 AstraClaude Instant 1.2vsGemini 4 HydrogenDeepSeek Coder V2-0724vsGemini 4 HydrogenClaude Instant 1.2vsMuse EmberDeepSeek Coder V2-0724vsMuse EmberClaude Instant 1.2vsGrok 4.7DeepSeek Coder V2-0724vsGrok 4.7Claude Instant 1.2vsMistral Large 4DeepSeek Coder V2-0724vsMistral Large 4Claude Instant 1.2vsKimi K3.1 ProDeepSeek Coder V2-0724vsKimi K3.1 ProFrequently asked questions
DeepSeek Coder V2-0724 leads Claude Instant 1.2 on 1 of the 1 benchmark they both report (HumanEval). Claude Instant 1.2 shipped 350 days before DeepSeek Coder V2-0724, so benchmark comparisons should account for the intervening progress.
Context windows are 100k (Claude Instant 1.2) vs 128k (DeepSeek Coder V2-0724). Claude Instant 1.2 is closed, while DeepSeek Coder V2-0724 is open weight.
On HumanEval, DeepSeek Coder V2-0724 leads at 87.2% vs Claude Instant 1.2 at 58.7%.