DeepSeek: DeepSeek V4.1 Flash
by DeepSeekDeepSeek: DeepSeek V4.1 Flash is a large language model from DeepSeek. It costs $0.045 per million input tokens and $0.500 per million output tokens. Its context window is 1.0M tokens.
- Input / 1M tokens
- $0.045
- Output / 1M tokens
- $0.500
- Cached input / 1M
- $0.0060
- Context window
- 1.0M
On repeated prefixes
tokens
Who serves it cheapest
32 hosts serve DeepSeek V4.1 Flash. Same weights, same API - the price difference is pure margin and routing.
| Provider | Input / 1M | Output / 1M | Context | Throughput | Uptime 24h |
|---|---|---|---|---|---|
| BaiduCheapestfp8 | $0.045 | $0.179 | 1.0M | - | 100.0% |
| Decartfp4 | $0.090 | $0.180 | 1.0M | - | 100.0% |
| Morphfp8 | $0.044 | $0.383 | 1.0M | - | 99.8% |
| Sail Researchfp4 | $0.090 | $0.250 | 1.0M | - | 100.0% |
| Wafermxfp4 | $0.045 | $0.500 | 1.0M | - | 100.0% |
| DeepInfrafp8 | $0.112 | $0.336 | 1.0M | - | 96.0% |
| Relace | $0.070 | $0.500 | 1.0M | - | 99.9% |
| InferenceNetfp8 | $0.103 | $0.600 | 1.0M | - | 100.0% |
| AtlasCloudfp8 | $0.135 | $0.540 | 1.0M | - | 98.4% |
| StreamLakefp8 | $0.147 | $0.588 | 1.0M | - | 100.0% |
| CoreWeavefp8 | $0.130 | $0.650 | 1.0M | - | 99.9% |
| Reka | $0.105 | $0.840 | 1.0M | - | 99.0% |
| GMICloudfp8 | $0.180 | $0.720 | 1.0M | - | 100.0% |
| Novitafp8 | $0.195 | $0.780 | 1.0M | - | 100.0% |
| Fireworks | $0.210 | $0.840 | 1.0M | - | 99.9% |
| Phala | $0.210 | $0.840 | 1.0M | - | 99.5% |
| DigitalOcean | $0.225 | $0.900 | 1.0M | - | 100.0% |
| Mancer 2fp8 | $0.200 | $1.00 | 1.0M | - | 100.0% |
| OpenInferencefp4 | $0.019 | $1.80 | 1.0M | - | 100.0% |
| Makorafp8 | $0.260 | $1.17 | 1.0M | - | 100.0% |
| Crusoefp8 | $0.290 | $1.20 | 1.0M | - | 100.0% |
| $0.300 | $1.20 | 1.0M | - | 99.1% | |
| Nebiusfp8 | $0.300 | $1.20 | 1.0M | - | 99.9% |
| PrimeIntellect | $0.300 | $1.20 | 1.0M | - | 99.9% |
| Cohere | $0.300 | $1.20 | 1.0M | - | 99.8% |
| BaseTen | $0.300 | $1.20 | 1.0M | - | 99.9% |
| Together | $0.300 | $1.20 | 1.0M | - | 100.0% |
| SiliconFlowfp8 | $0.300 | $1.20 | 1.0M | - | 100.0% |
| Modal | $0.300 | $1.20 | 1.0M | - | 100.0% |
| Parasailfp8 | $0.300 | $1.20 | 1.0M | - | 100.0% |
| Venicefp8 | $0.300 | $1.20 | 1M | - | 99.9% |
| Alibaba | $0.330 | $1.32 | 1M | - | 100.0% |
The spread between Baidu and Alibaba is 7.4× for identical weights. Quantization and context limits differ, so check both columns before switching.
About 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...
Specifications
| Model ID | deepseek/deepseek-v4.1-flash |
|---|---|
| Provider | DeepSeek |
| Context window | 1.0M tokens |
| Max output | 944K tokens |
| Input modalities | text, image |
| Output modalities | text |
| Knowledge cutoff | - |
| Open weights | Yes - deepseek-ai/DeepSeek-V4.1-Flash |
| Released | September 10, 2026 |
Cheaper alternatives
Models that cost less than DeepSeek V4.1 Flash while keeping at least half its context window and every input modality it supports.
Frequently asked
How much does DeepSeek: DeepSeek V4.1 Flash cost?
$0.045 per million input tokens and $0.500 per million output tokens. Cached input reads cost $0.0060 per million tokens.
What is the context window of DeepSeek: DeepSeek V4.1 Flash?
1.0M tokens, with up to 944K tokens of output per request.
Which provider serves DeepSeek: DeepSeek V4.1 Flash cheapest?
Baidu at $0.045 per million input tokens - 7.4× cheaper than Alibaba, the most expensive of the 32 hosts serving it.
Confirm against the source: DeepSeek official pricing.