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. | 128k | 1M |
| 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. | 34.38% | — |
MATH-500Reasoning — accuracy on the 500-problem MATH subset | 82.8% | — |
| Overview | ||
| Company | DeepSeek | Qwen |
| Release date | Dec 10 2024 | Aug 1 2025 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
DeepSeek V2.5 (Revised)vsClaude Fable 5.2Qwen3-FlashvsClaude Fable 5.2DeepSeek V2.5 (Revised)vsGPT-6.2 AstraQwen3-FlashvsGPT-6.2 AstraDeepSeek V2.5 (Revised)vsGemini 4 NeonQwen3-FlashvsGemini 4 NeonDeepSeek V2.5 (Revised)vsMuse EmberQwen3-FlashvsMuse EmberDeepSeek V2.5 (Revised)vsGrok 4.7Qwen3-FlashvsGrok 4.7DeepSeek V2.5 (Revised)vsMistral Large 4Qwen3-FlashvsMistral Large 4Frequently asked questions
DeepSeek V2.5 (Revised) and Qwen3-Flash don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. DeepSeek V2.5 (Revised) shipped 234 days before Qwen3-Flash, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (DeepSeek V2.5 (Revised)) vs 1M (Qwen3-Flash). DeepSeek V2.5 (Revised) is open weight, while Qwen3-Flash is closed.
Direct benchmark comparisons are unavailable — DeepSeek V2.5 (Revised) and Qwen3-Flash don't publish scores on any of the same benchmarks.