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. | 284B | 7B/14B |
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. | 1M | 1M |
| 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. | $0.0383Relace | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.0806Baidu | — |
| $12.09Baidu · 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. | 91.6% | 38.6% |
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. | 18% | — |
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. | — | 88.4% |
IFEvalAgentic tasks — following verifiable instructions in generated responses | — | 84.3% |
MATHReasoning — accuracy on the original competition mathematics test set | — | 79.5% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 88.1% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 63.3% |
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. | — | 94.8% |
| Overview | ||
| Company | DeepSeek | Qwen |
| Release date | Apr 24 2026 | Jan 27 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek-V4-FlashvsClaude Fable 5.2Qwen2.5-1MvsClaude Fable 5.2DeepSeek-V4-FlashvsGPT-6.2 AstraQwen2.5-1MvsGPT-6.2 AstraDeepSeek-V4-FlashvsGemini 4 NeonQwen2.5-1MvsGemini 4 NeonDeepSeek-V4-FlashvsMuse EmberQwen2.5-1MvsMuse EmberDeepSeek-V4-FlashvsGrok 4.7Qwen2.5-1MvsGrok 4.7DeepSeek-V4-FlashvsMistral Large 4Qwen2.5-1MvsMistral Large 4Frequently asked questions
DeepSeek-V4-Flash leads Qwen2.5-1M on 1 of the 1 benchmark they both report (LiveCodeBench). Qwen2.5-1M shipped 452 days before DeepSeek-V4-Flash, so benchmark comparisons should account for the intervening progress.
DeepSeek-V4-Flash has 284B parameters, while Qwen2.5-1M has 7B/14B. Context windows are 1M (DeepSeek-V4-Flash) vs 1M (Qwen2.5-1M).
On LiveCodeBench, DeepSeek-V4-Flash leads at 91.6% vs Qwen2.5-1M at 38.6%.