DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
Preise werden automatisch aktualisiert — täglich geprüft anhand der Listenpreise der Anbieter.
Independent benchmark scores for DeepSeek V4.1 Flash, measured by Artificial Analysis. Higher is better. Messmodus: Reasoning, Max Effort.
DeepSeek V4.1 Flash is an open-weight AI model by DeepSeek. You can download and self-host it for free; the prices below are hosted-API list prices, tracked daily, for when you prefer convenience over self-hosting.
DeepSeek V4.1 Flash ist ein KI-Sprachmodell von DeepSeek. Es ist open-weight: Sie können es herunterladen und auf Ihrer eigenen Hardware kostenlos ausführen. Es erzielt 39.5 im Artificial Analysis Intelligenzindex.
The weights are free and open — you can self-host DeepSeek V4.1 Flash and pay nothing per token. If you prefer a hosted API, list prices are $0.3 per million input tokens and $1.2 per million output tokens.
Unabhängige Benchmarks von Artificial Analysis geben ihm Humanity's Last Exam 39.2%, Long Context Reasoning 84%, SciCode 51.9%.
Es generiert etwa 219.8 Tokens pro Sekunde, mit einer Median 0.84s Verzögerung vor dem ersten Token. Unabhängig von Artificial Analysis gemessen.
Yes. DeepSeek V4.1 Flash has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.