DeepSeek V4 Flash 0731
deepseek/deepseek-v4-flash-0731
Overview
Capabilities
Tool callingVisionStructured outputStreaming
Pricing (lowest)
$0.035/M / $0.07/M
Context window
1M
Modality
Text
Hosts
6 hosts
Vendor details
| DeepSeek | Particle | Together AI | Baseten | Makora | Empiriolabs | |
|---|---|---|---|---|---|---|
| Input cost | $0.22/M | $0.035/M | $0.14/M | $0.13/M | $0.09/M | $0.14/M |
| Output cost | $0.66/M | $0.07/M | $0.28/M | $0.26/M | $0.195/M | $0.28/M |
| Cache read cost | $0.007/M | $0.007/M | $0.03/M | $0.028/M | $0.0196/M | - |
| Peak hours | 01:00–04:00, 06:00–10:00 UTC | Not time-varying | Not time-varying | Not time-varying | Not time-varying | Not time-varying |
| Peak input cost | $0.44/M | $0.035/M | $0.14/M | $0.13/M | $0.09/M | $0.14/M |
| Peak output cost | $1.32/M | $0.07/M | $0.28/M | $0.26/M | $0.195/M | $0.28/M |
| Peak cache read cost | $0.014/M | $0.007/M | $0.03/M | $0.028/M | $0.0196/M | - |
| Currency | USD | USD | USD | USD | USD | USD |
| Availability | available | available | available | available | available | available |
| Context window | 1M | 1M | 1M | 1M | 1M | 1M |
| Max output | 384K | 384K | 384K | 384K | 384K | 393K |
| Zero data retention | No | Yes | Yes | Yes | Yes | No |
| Launched | Jul 2026 | Jul 2026 | Aug 2026 | Jul 2026 | Jul 2026 | Jul 2026 |
| Capabilities | Tool callingTool choiceStreaming | Tool callingTool choiceStructured outputStreaming | Tool callingStructured outputStreaming | Tool callingTool choiceStructured outputStreaming | Tool callingStructured outputStreaming | Tool callingTool choiceStreaming |
| Modalities | Input: TextOutput: TextOutput: Tool use | Input: TextOutput: TextOutput: Tool use | Input: TextOutput: TextOutput: Tool use | Input: TextOutput: TextOutput: Tool use | Input: TextOutput: TextOutput: Tool use | Input: TextOutput: TextOutput: Tool use |