DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.
- Context window: 1,048,576 tokens
- Input: text
- Output: text
- Pricing: $0.088606/M input tokens, $0.177212/M output tokens
Community sentiment
Negative (-80) — 0 positive, 3 negative, 0 neutral across 3 community samples from X and the web, as of 2026-09-04.
Observers report multiple price increases for DeepSeek V4 Flash 0423 on OpenRouter across several repricings, with reported hikes around 12% and 26% that raise both input and output costs. Commenters are flagging repeated monthly increases and higher per-request rates.
- “🚨 DeepSeek V4 Flash 0423 just repriced on OpenRouter. DeepSeek moved the model from $0.07014/M input, $0.14028/M output to $0.088606/M input, $0.177212/M output. Both input and output are up 26%. https://t.co/cgo1mHQ2dN” — X
- “🚨 Five models were repriced on OpenRouter. DeepSeek V4 Flash 0423 rose input +12%, output +12%, its 3rd hike this month. Qwen3.5 397B A17B rose input +41%, output +50%. GLM 4.6 rose input +28%, output +26%. Qwen3 14B moved from $0.12/M input, $0.24/M output to $0.2275/M input, h” — X
- “🚨 DeepSeek V4 Pro 0423 and DeepSeek V4 Flash 0423 repriced on OpenRouter. DeepSeek V4 Pro 0423 moved from $0.417252/M input, $0.834504/M output to $1.60/M input, $3.20/M output, input +283%, output +283%. DeepSeek V4 Flash 0423 moved from $0.07952/M input, $0.15904/M output to h” — X
View on OpenRouter.
Model data sourced from
OpenRouter.