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 | 6B |
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 | 8k |
| 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. | $0.25SiliconFlow | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $1.00SiliconFlow | — |
| $100SiliconFlow · fp8 | — | |
| These models have no shared benchmark scores. | ||
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. | 49.2% | — |
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. | — | 58.5% |
MATHReasoning — accuracy on the original competition mathematics test set | — | 25.7% |
HellaSwagReasoning — commonsense completion of short scenarios | — | 79.7% |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | 59.4% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 68.4% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 81.2% | — |
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. | — | 61.4% |
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. | — | 72.3% |
| Overview | ||
| Company | DeepSeek | Z.ai |
| Release date | Mar 24 2025 | Oct 27 2023 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek V3-0324vsClaude Fable 5.2ChatGLM3-6BvsClaude Fable 5.2DeepSeek V3-0324vsGPT-6.2 AstraChatGLM3-6BvsGPT-6.2 AstraDeepSeek V3-0324vsGemini 4 HydrogenChatGLM3-6BvsGemini 4 HydrogenDeepSeek V3-0324vsMuse EmberChatGLM3-6BvsMuse EmberDeepSeek V3-0324vsGrok 4.7ChatGLM3-6BvsGrok 4.7DeepSeek V3-0324vsMistral Large 4ChatGLM3-6BvsMistral Large 4Frequently asked questions
DeepSeek V3-0324 and ChatGLM3-6B don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. ChatGLM3-6B shipped 514 days before DeepSeek V3-0324, so benchmark comparisons should account for the intervening progress.
DeepSeek V3-0324 has 671B parameters, while ChatGLM3-6B has 6B. Context windows are 128k (DeepSeek V3-0324) vs 8k (ChatGLM3-6B).
Direct benchmark comparisons are unavailable — DeepSeek V3-0324 and ChatGLM3-6B don't publish scores on any of the same benchmarks.