DeepSeek
DeepSeek-V4-Flash
azure/deepseek-v4-flash
- Input: text
- Output: text
RuntimeWire coverage
- Perplexity launches a 190M-document retrieval test and keeps the answers private
- NVIDIA says new optimizations make local agents up to 1.9x faster
- DeepSeek publishes 168GB vision model weights under an MIT license
- Head to head: Google: Gemini 3.7 Flash vs DeepSeek-V4-Flash
- Ox Alpha filters domestic Chinese political risks, CTGT audit finds
- OneTriangle launches DeepSeek V4 Flash hosting at $0.15 per million input tokens
- Anonymous Ox Alpha processes 26T tokens on OpenCode, breaks OpenRouter launch record
- FreeToken reports 22-25 tok/s for a 284B DeepSeek model on one RTX 5090
- DeepSeek's experimental vision model spans three formats, caps images at 384 tokens
- Maple said it made DeepSeek V4 Flash available to paid customers
- Open-weight models surge past closed rivals in Vercel token traffic
- Nicholai Mitchko releases self-hosted DeepSeek latent-reasoning stack for Blackwell GPUs
View on OpenRouter. Model data sourced from OpenRouter.