DeepSeek V4 Flash 0423 vs DeepSeek V4 Flash 0731 (Thinking)
DeepSeek V4 Flash 0731 (Thinking) is cheaper on output tokens. Choose DeepSeek V4 Flash 0423 or DeepSeek V4 Flash 0731 (Thinking) based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | DeepSeek V4 Flash 0423 | DeepSeek V4 Flash 0731 (Thinking) |
|---|---|---|
| Provider | DeepSeek | NanoGPT |
| Input / 1M tokens | $0.09 | $0.05 |
| Output / 1M tokens | $0.18 | $0.16 |
| Context window | 1M | 1M |
| Parameters | 284B | — |
| Open weights | Yes | Yes |
| Released | Apr 2026 | Jul 2026 |
| GPQA Diamond score | 91.0 | — |
| LiveBench score | 65.5 | — |
| Epoch Capabilities Index score | 154 | — |
| LMArena (Chatbot Arena) Elo score | 1432 | — |
| LMArena WebDev Elo score | 1582 | — |
FAQ
Is DeepSeek V4 Flash 0423 or DeepSeek V4 Flash 0731 (Thinking) cheaper?
DeepSeek V4 Flash 0731 (Thinking) is cheaper on output tokens ($0.16 vs $0.18 per 1M).
Which has the larger context window, DeepSeek V4 Flash 0423 or DeepSeek V4 Flash 0731 (Thinking)?
Both models offer a similar context window.
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.