Home / DeepSeek / DeepSeek V4 Flash (latest)
DeepSeek V4 Flash (latest) API price
deepseek/deepseek-v4-flash
The cheapest DeepSeek V4 Flash (latest) API is $0.0826/M input at StreamLake (via OpenRouter) - all 20 providers compared below, per 1M tokens.
Rolling name - always points to the newest snapshot of this model. The pinned snapshots are priced separately: DeepSeek V4 Flash (0731 snapshot).
SmallDeepSeek1.0M context20 sellers
Input
$0.0826
per 1M tokens · StreamLake (via OpenRouter)
up to $0.44
Output
$0.1652
per 1M tokens · StreamLake (via OpenRouter)
up to $1.536
Where to buy it
Cheapest DeepSeek V4 Flash (latest) API providers compared - 20 sellers, from $0.0826/M.
Cost calculator
| Provider | Input $/M | Output $/M | Cached in | Uptime | Your cost |
|---|---|---|---|---|---|
| StreamLake (via OpenRouter)fp8 | 0.083cheapest | 0.165 | 0.017 | 100% | |
| DeepInfrafp8 | 0.090 | 0.180 | 0.018 | 100% | |
| GMICloud (via OpenRouter)fp8 | 0.091 | 0.182 | 0.018 | 100% | |
| Venice (via OpenRouter) | 0.097 | 0.193 | 0.020 | 100% | |
| DigitalOcean (via OpenRouter) | 0.098 | 0.196 | 0.020 | 100% | |
| SiliconFlow (via OpenRouter)fp8 | 0.130 | 0.280 | 0.028 | 100% | |
| Venice | 0.138 | 0.275 | 0.028 | 100% | |
| AtlasCloudfp4 | 0.140 | 0.280 | 0.028 | 100% | |
| Baidu (via OpenRouter)fp8 | 0.140 | 0.280 | 0.028 | 100% | |
| Parasailfp8 | 0.140 | 0.280 | 0.070 | 100% | |
| Alibaba (via OpenRouter)fp8 | 0.150 | 0.300 | 0.030 | 100% | |
| Wafer (via OpenRouter) | 0.400 | 0.170 | 0.013 | 100% | |
| Mancer 2 (via OpenRouter)fp8 | 0.190 | 0.500 | - | 96.7% | |
| Azure (via OpenRouter) (us) | 0.210 | 0.560 | 0.031 | 100% | |
| OrcaRouter (aggregator) | 0.220 | 0.660 | - | - | |
| Cheaper Inference (discount marketplace)40% off | 0.180 | 0.720 | 0.0036 | - | |
| Relace (via OpenRouter)fp4 | 0.0099 | 1.28 | 0.0099 | 100% | |
| OpenInference (via OpenRouter)fp4 | 0.010 | 1.54 | 0.010 | 100% | |
| Cloudflare (via OpenRouter) | 0.440 | 1.32 | 0.014 | 80% | |
| TokenRouter (aggregator) | 0.440 | 1.32 | 0.014 | - |
44.4× between the cheapest and the dearest of 20 sellers.
Against its tier
- This model$0.2478
- Cheapest · Nemotron 3.5 Lightning$0.145
- Tier median · Qwen 3.8 Omni Flash$0.62
- Dearest · GPT 5.4 Mini (03-17 snapshot)$5.25
#3 of 19 small · cheaper than 89%
Details
- Context window
- 1.0M1,048,576 tokens
- Developer
- DeepSeek
- Sellers
- 2044.4× spread
- Cheapest at
- StreamLake (via OpenRouter)
- Cached input
- $0.0165where the source publishes one
- Combined
- $0.2478in + out per 1M
- Last checked
- 2026-10-12
- Availability
- 100.00%median of 16 OpenRouter endpoints, last 30 min