DeepSeek V4 Flash 0423
Provider: DeepSeek CN
Overview
DeepSeek V4 Flash 0423 is a budget-tier Mixture-of-Experts model featuring 284B total parameters with 13B activated. It supports 1M-token context and large outputs for high-throughput inference workloads.
Strengths
- 284B parameters with 13B activated
- 1,048,576 token context window
- 393,216 max output tokens
- Function calling and JSON mode supported
Best use cases
- Processing large multi-megabyte codebases
- Extracting structured data via JSON mode
- Long-document retrieval and summarization
- High-volume API integration pipelines
Pricing position
Input costs $0.105/M and output $0.21/M tokens, undercutting the V4 Flash 0731 sibling and matching GPT-4.1 Nano input pricing.
Caveats
No free tier is available on this model.
Notes drafted by an LLM from the figures on this page (2026-11-14) and rewritten when the price changes.
Key facts
- Price basis
- Median third-party host price (no first-party API)
- Max output
- 393K tokens
- Listed on OpenRouter
- 2026-04-24
- Tracked since
- 2026-08-04
- Last updated
- 2026-11-14
Features
Price history
- 2026-11-14 Price change $0.05 / $0.45 → $0.10 / $0.21
- 2026-11-13 Price change $0.12 / $0.17 → $0.05 / $0.45
- 2026-11-12 Price change $0.05 / $0.45 → $0.12 / $0.17
- 2026-11-12 Price change $0.12 / $0.17 → $0.05 / $0.45
- 2026-11-10 Price change $0.05 / $0.45 → $0.12 / $0.17
Monthly cost estimate
| Scale | Input (M) | Output (M) | Monthly |
|---|---|---|---|
| 0.1M (hobby) | 0.1 | 0.05 | $0.02 |
| 1M (small app) | 1 | 0.5 | $0.21 |
| 10M (mid scale) | 10 | 5 | $2.10 |
| 100M (large) | 100 | 50 | $21.00 |
※ Cache and batch discounts not factored in. Detailed calculator
Official resources
Other models from DeepSeek
Similar-priced models
Source: OpenRouter API (deepseek/deepseek-v4-flash)