DeepSeek V4-Flash
Key facts · verified Oct 29, 2026
DeepSeek V4-Flash is a DeepSeek model released on 31 Jul 2026. It was retired on 10 September 2026. Its last listed API price was $0.15 per million input tokens and $0.60 per million output tokens (off-peak; $0.30 / $1.20 peak — figures are DeepSeek-V4.1-Flash, which now serves this model). Its context window is 1M tokens. It ranks #104 on LMArena's text leaderboard (Elo 1436, read 1 Nov 2026). Source: api-docs.deepseek.com.
Retired on 10 September 2026. DeepSeek withdrew V4-Flash and now serve requests to deepseek-v4-flash with DeepSeek-V4.1-Flash, published as deepseek-flash. The row is kept so that links and comparisons pointing at it still resolve; the figures shown are V4.1-Flash, which is what an API call to the old name returns today.
Key Features
Full Specifications
$0.15 / $0.60 per million tokens | |
|---|---|
| Identity | |
| Developer | DeepSeek |
| Released | Jul 2026 |
| Retired | 10 September 2026 |
| Status | Retired. DeepSeek retired V4-Flash on 2026-09-10; requests to deepseek-v4-flash are now served by DeepSeek V4.1-Flash, which has its own page. The figures below are V4.1-Flash's (off-peak), recorded here before that page existed. |
| Licence | MIT |
| Self-hostable | Yes - open weights on Hugging Face |
| Cost | |
| Blended $/1M tokensinput × 0.75 + output × 0.25 | $0.262 / 1M tokens |
| Input price | $0.150 / 1M tokens |
| Output price | $0.600 / 1M tokens |
| Cached input | $0.006 / 1M tokens (peak); $0.003 off-peak - V4.1-Flash rates, which now serve this name |
| Batch discount | Off-peak billing: 50% off all rates outside 01:00-04:00 and 06:00-10:00 UTC, Monday to Friday (weekends and Chinese public holidays are off-peak in full) |
| Free tier | Not offered |
| Capacity | |
| Context window | 1M tokens |
| Max output | 384k tokens |
| Long-context surcharge | None — flat rate across the 1M window |
| Capability | |
| Vision in | No |
| Audio in | No |
| Function calling | Yes |
| Structured output | Yes — JSON output |
| Extended reasoning | Yes — thinking (default) and non-thinking modes |
| Web search | No |
| Code execution | No |
| Access | |
| API | Yes |
| Chat app | Yes |
| Cloud marketplaces | Not offered |
| Fine-tuning | Not offered |
| Measured quality | |
| LMArena Elo | 1436 (checked Nov 2026) |
| LMArena rank | Rank 104 |
| Elo per dollarLMArena Elo ÷ blended $/1M tokens | 5470 |