DeepSeekopen weightsDeepSeek V4 Flash 0731 pricing
$0.10 per 1M input tokens and $0.30 per 1M output tokens. In real work that is 0.19¢ per chat answer and 0.19¢ to summarize a 20-page PDF.
What DeepSeek V4 Flash 0731 costs per task
Token counts estimated from a measured model of the same lab (DeepSeek V4.1 Flash), rescaled by tokenizer. Reasoning tokens are billed as output.
| Task | Cost | Tokens used | Price rank | Basis |
|---|---|---|---|---|
| Answer a chat question | 0.19¢$1.90 / 1,000 | 65 in · 859 out · 5,461 reasoning | 153 of 336 | estimated |
| Reply to a customer email | 0.06¢$0.62 / 1,000 | 279 in · 225 out · 1,749 reasoning | 123 of 336 | estimated |
| Write a 1,500-word blog post | 0.09¢$0.92 / 1,000 | 81 in · 2,230 out · 809 reasoning | 20 of 336 | estimated |
| Write 10 product descriptions | 0.08¢$0.84 / 1,000 | 307 in · 870 out · 1,842 reasoning | 29 of 336 | estimated |
| Translate 1,000 words | 0.36¢$3.65 / 1,000 | 1,354 in · 1,417 out · 10,291 reasoning | 130 of 336 | estimated |
| Proofread one page (500 words) | 0.09¢$0.86 / 1,000 | 746 in · 660 out · 1,954 reasoning | 85 of 336 | estimated |
| Summarize a 20-page PDF | 0.19¢$1.89 / 1,000 | 12,970 in · 427 out · 1,565 reasoning | 66 of 336 | estimated |
| Extract data from an invoice | 0.09¢$0.91 / 1,000 | 395 in · 449 out · 2,436 reasoning | 92 of 336 | estimated |
| Analyze a 200-row spreadsheet | ≥0.53¢$5.28 / 1,000 | 4,801 in · 0 out · 16,000 reasoning | 72 of 336 | estimated |
| Answer from 5 retrieved documents (RAG) | 0.15¢$1.49 / 1,000 | 3,696 in · 648 out · 3,102 reasoning | 85 of 336 | estimated |
| Classify 100 support tickets | 0.16¢$1.61 / 1,000 | 1,296 in · 461 out · 4,462 reasoning | 55 of 336 | estimated |
| Review 500 lines of code | ≥0.53¢$5.28 / 1,000 | 4,801 in · 0 out · 16,000 reasoning | 93 of 336 | estimated |
| Write a SQL query | 0.07¢$0.65 / 1,000 | 145 in · 406 out · 1,723 reasoning | 64 of 336 | estimated |
| Run a 50-step coding agent | 10¢$103 / 1,000 | 2,235,262 in · 14,985 out · 87,450 reasoning | 57 of 336 | modeled |
Price at every provider
List prices per 1M tokens (input / output) at each host serving DeepSeek V4 Flash 0731 through OpenRouter today. Hosts can differ in quantization, context and speed.
| Provider | Input / output | Cached input | Context | Uptime |
|---|---|---|---|---|
| O OpenInferencefp4 | $0.012 / $0.14 | $0.002 | 1M | 100% |
| B Baidufp8 | $0.06 / $0.18 | $0.002 | 1M | 100% |
 DeepInfrafp8 | $0.06 / $0.18 | $0.015 | 1M | 100% |
| W Wafernvfp4 | $0.13 / $0.23 | $0.12 | 1M | 100% |
| D DigitalOceanunknown | $0.12 / $0.24 | $0.024 | 1M | 100% |
| S StreamLakefp8 | $0.088 / $0.26 | $0.003 | 1M | 100% |
| B BaseTenfp8 | $0.13 / $0.26 | $0.028 | 1M | 100% |
| B BaseTenfp32 | $0.13 / $0.26 | $0.028 | 1M | 100% |
| V Veniceunknown | $0.13 / $0.26 | $0.028 | 1M | 100% |
| C CoreWeavefp8 | $0.13 / $0.28 | $0.07 | 262K | 100% |
 Cohereunknown | $0.14 / $0.28 | $0.07 | 1M | 100% |
 Togetherunknown | $0.14 / $0.28 | $0.03 | 1M | 100% |
 Parasailfp8 | $0.14 / $0.28 | $0.05 | 1M | 100% |
| S Sail Researchfp4 | $0.10 / $0.30 | $0.03 | 1M | 100% |
| S Sail Researchfp4 | $0.10 / $0.30 | $0.03 | 1M | 100% |
| I Inceptronfp4 | $0.13 / $0.48 | $0.027 | 1M | 100% |
| R Rekaunknown | $0.021 / $0.53 | $0.006 | 262K | 100% |
| M Mancer 2fp8 | $0.20 / $0.60 | - | 1M | 100% |
 SiliconFlowfp8 | $0.22 / $0.66 | $0.028 | 1M | 100% |
| G GMICloudfp8 | $0.29 / $0.86 | $0.009 | 1M | 100% |
 Alibabaunknown | $0.35 / $1.06 | $0.035 | 1M | 93% |
Alibabaunknown | $0.37 / $1.12 | $0.037 | 1M | |
| R Relacefp4 | $0.015 / $1.28 | $0.015 | 1M | 100% |
 Novitafp8 | $0.41 / $1.23 | $0.026 | 1M | 100% |
| P Phalaunknown | $0.44 / $1.32 | $0.028 | 1M | 100% |
| A AtlasCloudfp4 | $0.44 / $1.32 | $0.028 | 1M | 100% |
| C Cloudflareunknown | $0.44 / $1.32 | $0.014 | 1M | 100% |
Price history
First seen in our price history on Aug 3, 2026 at $0.09 input / $0.18 output. Since then the output price moved from $0.18 to $0.30 per 1M tokens (12 changes).
FAQ
How much does DeepSeek V4 Flash 0731 cost?
DeepSeek V4 Flash 0731 costs $0.10 per 1 million input tokens and $0.30 per 1 million output tokens, with cached input at $0.03 (list price on Oct 18, 2026). In practice one chat answer costs about 0.19¢ and summarizing a 20-page PDF about 0.19¢.
Where is DeepSeek V4 Flash 0731 cheapest?
Among 27 providers we track, the lowest list price today is at OpenInference: $0.012 input and $0.14 output per 1M tokens.
Does DeepSeek V4 Flash 0731 charge for reasoning tokens?
Yes. Reasoning (thinking) tokens are billed as output tokens. That is why the cost per task can be much higher than the visible answer suggests - our task costs include them.
Is the DeepSeek V4 Flash 0731 cost per task measured or estimated?
Estimated: DeepSeek V4 Flash 0731 was not run directly. Tokens come from DeepSeek V4.1 Flash and are rescaled by the tokenizer of DeepSeek. Real usage can differ.