DeepSeek V4.1 Flash
by DeepSeek · deepseek/deepseek-v4.1-flash
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...
- Input / 1M tokens
- $0.045
- Output / 1M tokens
- $0.50
- Context
- 1.05M
Specs
| Input price | $0.045 / 1M tokens |
|---|---|
| Output price | $0.50 / 1M tokens |
| Cached input | $0.006 / 1M tokens |
| Context window | 1.05M tokens |
| Max output | 944K tokens |
| Reads | text, image |
| Writes | text |
| Tool calling | Yes |
| Reasoning | Yes |
| Listed | Sep 10, 2026 |
List prices in USD, updated Oct 30, 2026. Providers may charge differently.
Providers
41 providers serve this model, as of Oct 30, 2026, 18:12 UTC.
| Provider | Quantization | Context | Max output | Input / 1M | Output / 1M | Cache read / 1M | Uptime (30 min) | Latency (median) | Throughput (median) | Supports | Status |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Morph | fp8 | 1.05M | 944K | $0.022 | $0.192 | $0.004 | 100% | 674 ms | 72 tok/s | Normal | |
| Decart | fp4 | 1.05M | 262K | $0.09 | $0.18 | $0.018 | 100% | 668 ms | 79 tok/s | Normal | |
| Sail Research | fp4 | 1.05M | 384K | $0.09 | $0.25 | $0.006 | 100% | 544 ms | 50 tok/s | Normal | |
| Sail Research | fp4 | 1.05M | 384K | $0.09 | $0.25 | $0.006 | 100% | 492 ms | 48 tok/s | Normal | |
| DeepInfra | fp8 | 1.05M | 131K | $0.112 | $0.336 | $0.00336 | 100% | 1.96 s | 74 tok/s | Normal | |
| Wafer | mxfp4 | 1.05M | 944K | $0.045 | $0.50 | $0.006 | 100% | 349 ms | 153 tok/s | Normal | |
| DeepInfra | fp8 | 1.05M | 131K | $0.14 | $0.42 | $0.0042 | 100% | 611 ms | 73 tok/s | Normal | |
| Relace | — | 1.05M | 944K | $0.07 | $0.50 | $0.008 | 100% | 1.03 s | 66 tok/s | Normal | |
| AtlasCloud | fp8 | 1.05M | 393K | $0.135 | $0.54 | $0.0135 | 100% | 736 ms | 91 tok/s | Normal | |
| StreamLake | fp8 | 1.02M | 384K | $0.147 | $0.588 | $0.00294 | 100% | 1.22 s | 78 tok/s | Normal | |
| InferenceNet | fp8 | 1.04M | 393K | $0.174 | $0.586 | $0.006 | 100% | 396 ms | 82 tok/s | Normal | |
| InferenceNet | fp8 | 1.04M | 393K | $0.174 | $0.586 | $0.006 | 100% | 369 ms | 119 tok/s | Normal | |
| CoreWeave | fp8 | 1.05M | 944K | $0.13 | $0.65 | $0.002 | 100% | 515 ms | 205 tok/s | Normal | |
| Alibaba | — | 1M | 393K | $0.165 | $0.66 | $0.0165 | 100% | 719 ms | 126 tok/s | Normal | |
| DeepInfra | fp8 | 1.05M | 131K | $0.21 | $0.63 | $0.0063 | 0% | 450 ms | 95 tok/s | Code -5 | |
| Wafer | mxfp4 | 1.05M | 944K | $0.19 | $0.65 | $0.012 | 100% | 332 ms | 163 tok/s | Normal | |
| GMICloud | fp8 | 1.05M | 944K | $0.18 | $0.72 | $0.0036 | 100% | 1.68 s | 85 tok/s | Normal | |
| Reka | — | 1.05M | 393K | $0.105 | $0.84 | $0.007 | 100% | 333 ms | 80 tok/s | Normal | |
| Novita | fp8 | 1.05M | 393K | $0.195 | $0.78 | $0.0039 | 100% | 952 ms | 107 tok/s | Normal | |
| Fireworks | — | 1.05M | 944K | $0.21 | $0.84 | $0.0042 | 100% | 1.50 s | 97 tok/s | Normal | |
| Phala | — | 1.05M | 944K | $0.21 | $0.84 | $0.0042 | 100% | 1.11 s | 95 tok/s | Normal | |
| OpenInference | fp4 | 1.05M | 944K | $0.0433 | $1.04 | $0.0399 | 100% | 637 ms | 100 tok/s | Normal | |
| DigitalOcean | — | 1.05M | 944K | $0.225 | $0.90 | $0.006 | 100% | 656 ms | 154 tok/s | Normal | |
| Mancer 2 | fp8 | 1.05M | 944K | $0.20 | $1.00 | — | 100% | 728 ms | 40 tok/s | Normal | |
| Makora | fp8 | 1.05M | 944K | $0.26 | $1.17 | $0.0058 | 100% | 358 ms | 106 tok/s | Normal | |
| Crusoe | fp8 | 1.05M | 944K | $0.29 | $1.20 | $0.007 | 100% | 622 ms | 59 tok/s | Normal | |
| Baidu | fp8 | 1.05M | 393K | $0.30 | $1.20 | $0.006 | 100% | 932 ms | 82 tok/s | Normal | |
| BaseTen | — | 1.05M | 262K | $0.30 | $1.20 | $0.007 | 100% | 211 ms | 156 tok/s | Normal | |
| BaseTen | — | 1.05M | 262K | $0.30 | $1.20 | $0.007 | 100% | 229 ms | 170 tok/s | Normal | |
| Cohere | — | 1.05M | 384K | $0.30 | $1.20 | $0.003 | 100% | 346 ms | 173 tok/s | Normal | |
| Fireworks | — | 1.05M | 944K | $0.30 | $1.20 | $0.006 | 100% | 557 ms | 73 tok/s | Normal | |
| — | 1.05M | 944K | $0.30 | $1.20 | $0.03 | 100% | 911 ms | 41 tok/s | Normal | ||
| Modal | — | 1.05M | 944K | $0.30 | $1.20 | $0.03 | 100% | 372 ms | 100 tok/s | Normal | |
| Nebius | fp8 | 1.05M | 944K | $0.30 | $1.20 | $0.01 | 100% | 511 ms | 107 tok/s | Normal | |
| Parasail | fp8 | 1.05M | 944K | $0.30 | $1.20 | $0.006 | 100% | 421 ms | 189 tok/s | Normal | |
| PrimeIntellect | — | 1.05M | 131K | $0.30 | $1.20 | $0.006 | 100% | 538 ms | 148 tok/s | Normal | |
| SiliconFlow | fp8 | 1.05M | 393K | $0.30 | $1.20 | $0.006 | 100% | 1.38 s | 93 tok/s | Normal | |
| Together | — | 1.05M | 944K | $0.30 | $1.20 | $0.006 | 100% | 193 ms | 214 tok/s | Normal | |
| Venice | fp8 | 1M | 131K | $0.30 | $1.20 | $0.0075 | 100% | 435 ms | 171 tok/s | Normal | |
| Fireworks | — | 1.05M | 944K | $0.45 | $1.80 | $0.009 | 100% | 333 ms | 232 tok/s | Normal | |
| BaseTen | — | 1.05M | 131K | $0.60 | $2.40 | $0.014 | 16.7% | 318 ms | 233 tok/s | Code -5 |
"Cheapest" is the lowest input plus output price; "Fastest" the highest median throughput. Columns empty for every provider are not shown.
Usage
#4
by tokens, Oct 29, 2026
2.23T
tokens that day · 7 days −32%
Data as of 2026-10-30T18:12:03.879Z.All rankings
Related models
- DeepSeek
DeepSeek V4.1 Flash 1017
- Input
- $0.15
- Output
- $0.60
- Context
- 1.05M
- DeepSeek
DeepSeek V4 Flash 0731
- Input
- $0.119
- Output
- $0.238
- Context
- 1.05M
- DeepSeek
DeepSeek V4 Flash 0423
- Input
- $0.091
- Output
- $0.182
- Context
- 1.05M
- DeepSeek
DeepSeek V4 Pro 0813
- Input
- $0.66
- Output
- $1.98
- Context
- 1.05M
- DeepSeek
DeepSeek V4 Pro 0423
- Input
- $0.948
- Output
- $1.90
- Context
- 1.05M
- DeepSeek
DeepSeek V4.1 Flash (batch)
- Input
- $0.05
- Output
- $0.15
- Context
- 1.05M