DeepSeek V4 Flash 0731 API pricing
DeepSeek · Budget price tier · MIT · released 2026-07. Prices in USD per 1 million tokens. Data snapshot: 2026-11-12.
A support ticket workload of 100K requests a month costs $21.51 on DeepSeek V4 Flash 0731; the next cheaper model in the Budget price tier for this workload, Nova Lite 1.0, does it for $21.00. The price we quote for DeepSeek V4 Flash 0731 has changed 36 times since 2026-08-15, most recently by -2.1% on 2026-11-08; it is now the 47th cheapest of 160 models in the Budget price tier. Buying from Relace rather than at the price we quote lowers the token price by about 47%, for a quantized fp4 build. No Arena score is published for it, so it carries no Efficiency score and cannot hold the FinOps Friendly badge.
- Input: $0.0434 per 1M tokens
- Output: $0.30 per 1M tokens
- Batch API (in/out): Not published
- Prompt-cache read: $0.035 per 1M tokens
- Context window: 1M tokens
- Arena ELO: Not published
Available on
Direct API, OpenRouter
Curated by hand; check with the provider before committing.
Market position
On blended price ($0.223 per 1M tokens at 30% input, 70% output) it is cheaper than 71% of models in the Budget price tier: 47th cheapest of 160. Across every price tier it is cheaper than 85% of priced text models in the catalogue: 47th cheapest of 303. No Arena score has been published for DeepSeek V4 Flash 0731, so this page does not rank it on quality and OptimNow's quality bars do not apply to it. 77 of the 160 models we list in the Budget price tier carry one. An unranked model is unmeasured, not poor: the benchmark needs votes, and a recent release has not had time to collect them.
Licence, size and capabilities
DeepSeek V4 Flash 0731 is open source: its weights are published under MIT, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token. Its published size is 284B parameters. Listed capabilities: Text, Reasoning, Agents. Filling its 1M context window once costs $0.0434 in input tokens at list price.
Pricing by host
A host is a company that runs the model and sells it by the token: the model maker, a cloud platform like Amazon Bedrock, Azure or Vertex, or an independent host. One model, several hosts, and often several prices. This model is sold by 23 different hosts. The price we quote elsewhere on this page is marked. We quote one host's rate so that models stay comparable; it is not a recommendation to buy there.
Cheapest host: Relace at $0.07 input / $0.14 output per 1M tokens, 47% below the price we quote on the blended price (30% input, 70% output). It serves a quantized fp4 build, which is not the same product as the full-precision model. A support ticket workload at 100K requests a month costs $17.50 there against $21.51 at the price we quote, a difference of $4.01 a month. The dearest, Cloudflare, charges 8.9x the cheapest.
| Host | Input $/1M | Output $/1M | Precision |
|---|---|---|---|
| Relace | $0.07 | $0.14 | fp4 |
| DeepInfra | $0.06 | $0.18 | fp8 |
| DigitalOcean | $0.119 | $0.238 | not declared |
| Wafer | $0.14 | $0.23 | nvfp4 |
| BaseTen | $0.13 | $0.26 | fp8 |
| Venice | $0.13 | $0.26 | not declared |
| CoreWeave | $0.13 | $0.27 | fp8 |
| Cohere | $0.14 | $0.28 | fp8 |
| Parasail | $0.14 | $0.28 | fp8 |
| Sail Research | $0.15 | $0.30 | fp4 |
| StreamLake | $0.132 | $0.396 | fp8 |
| Reka | $0.10 | $0.42 | not declared |
| Inceptron | $0.13 | $0.495 | fp4 |
| Mancer 2 | $0.15 | $0.50 | fp8 |
| SiliconFlow | $0.22 | $0.66 | fp8 |
| GMICloud | $0.286 | $0.858 | fp8 |
| Novita | $0.308 | $0.924 | fp8 |
| Alibaba | $0.352 | $1.056 | not declared |
| Chutes | $0.44 | $1.32 | fp8 |
| PrimeIntellect | $0.44 | $1.32 | fp4 |
| Baidu | $0.44 | $1.32 | fp8 |
| AtlasCloud | $0.44 | $1.32 | fp4 |
| Cloudflare | $0.44 | $1.32 | not declared |
List prices as published on 2026-11-12, via OpenRouter. Negotiated rates, committed-use discounts and provisioned capacity are not shown. These hosts do not all serve the model at the same numeric precision, so the cheapest is not a like-for-like substitute for the dearest. These listings are promotional rates, below the host's standard rate: StreamLake's, 70% off; GMICloud's, 35% off; and Novita's, 30% off.
Monthly cost at 100K requests, by business use case
From the after-discounts figure (prompt caching on the cached input share, batch API for async workloads, only where DeepSeek publishes those rates), which assumes you implement both, to the list price. One figure means no discount is published.
| Use case | Tokens (in/out) | Monthly (after discounts – list) |
|---|---|---|
| Support Ticket | 1,500 / 500 | $21 – $22 |
| Knowledge Q&A | 2,000 / 800 | $32 – $33 |
| Meeting Summary | 10,000 / 1,200 | $79 |
| Marketing Content | 2,500 / 1,800 | $64 – $65 |
| Coding Task | 3,000 / 2,000 | $72 – $73 |
| Invoice Processing | 1,500 / 600 | $24 – $25 |
| Call Summary | 2,000 / 700 | $30 |
| Agent Workflow | 6,000 / 3,000 | $113 – $116 |
Price history
DeepSeek V4 Flash 0731 has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.14 input / $0.28 output per 1M tokens. The price we quote has changed 36 times since then, 5 of them a change of quoted host rather than a repricing. The latest change, on 2026-11-08, was -2.1% on the blended price (30% input, 70% output), to $0.0434 input / $0.30 output. Net of all changes, the blended price is -6.3% against 2026-08-15.
| Date | Old price (in / out) | New price (in / out) | Blended change |
|---|---|---|---|
| 2026-11-08 | $0.059 / $0.30 | $0.0434 / $0.30 | -2.1% |
| 2026-11-03 | $0.119 / $0.238 | $0.059 / $0.30 | +12.6% |
| 2026-10-27 | $0.0444 / $0.1333 | $0.119 / $0.238 | +89.7% (quoted host changed, not a repricing) |
| 2026-10-24 | $0.44 / $1.32 | $0.0444 / $0.1333 | -89.9% |
| 2026-10-20 | $0.15 / $0.15 | $0.44 / $1.32 | +604.0% (quoted host changed, not a repricing) |
31 earlier changes are not listed.
Alternatives to DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 carries no Arena score, so this page cannot place it against models of comparable quality. The alternatives below are its nearest neighbours on price in the Budget price tier.
- Next cheaper model in the Budget price tier: Qwen3 32B (Alibaba) at $0.22 blended, 1% less.
- Next more expensive: Hy-MT2-30B-A3B (Tencent) at $0.2287 blended, 3% more.
DeepSeek V4 Flash 0731 head to head
Other models from DeepSeek
The 8 nearest to DeepSeek V4 Flash 0731 in price, of 13 other DeepSeek models in the catalogue. Prices per 1M tokens, input then output.
- DeepSeek V4 Flash 0423: $0.09 / $0.18
- DeepSeek V4.1 Flash: $0.022 / $0.40
- DeepSeek V3.2: $0.26 / $0.38
- DeepSeek V3.2 Exp: $0.27 / $0.41
- DeepSeek V4.1 Flash 1017: $0.15 / $0.60
- DeepSeek V3: $0.32 / $0.89
- DeepSeek V3.1: $0.25 / $0.95
- DeepSeek V3.1 Terminus: $0.27 / $1.00
All 14 DeepSeek models and prices
Common questions about DeepSeek V4 Flash 0731 pricing
- How much does DeepSeek V4 Flash 0731 cost?
- We quote DeepSeek V4 Flash 0731 at $0.0434 per 1M input tokens and $0.30 per 1M output tokens. That figure matches none of the rates its 23 hosts list. Prices as listed on 2026-11-12.
- Which host is cheapest for DeepSeek V4 Flash 0731?
- Cheapest host: Relace at $0.07 input / $0.14 output per 1M tokens, 47% below the price we quote on the blended price (30% input, 70% output). It serves a quantized fp4 build, which is not the same product as the full-precision model. A support ticket workload at 100K requests a month costs $17.50 there against $21.51 at the price we quote, a difference of $4.01 a month. The dearest, Cloudflare, charges 8.9x the cheapest. Host rates as listed on 2026-11-12.
- Does DeepSeek V4 Flash 0731 offer batch or cached pricing?
- Yes, for prompt caching only. Cache reads cost $0.035 per 1M tokens, as listed on 2026-11-12. No batch pricing is published.
- What would DeepSeek V4 Flash 0731 cost per month?
- With the Support Ticket profile (1,500 input and 500 output tokens per request), 100,000 requests a month cost about $22 on DeepSeek V4 Flash 0731 at the list price quoted on 2026-11-12.
- Can DeepSeek V4 Flash 0731 be self-hosted?
- Yes. DeepSeek V4 Flash 0731 is open source: its weights are published under MIT, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token.
- Has the price of DeepSeek V4 Flash 0731 changed?
- DeepSeek V4 Flash 0731 has been in our daily price archive since tracking began on 2026-08-15, when it listed at $0.14 input / $0.28 output per 1M tokens. The price we quote has changed 36 times since then, 5 of them a change of quoted host rather than a repricing. The latest change, on 2026-11-08, was -2.1% on the blended price (30% input, 70% output), to $0.0434 input / $0.30 output. Net of all changes, the blended price is -6.3% against 2026-08-15.
Its weights are published, which changes what you may run yourself: the cost gap between open weights and proprietary APIs.
These are list prices. Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Build the business case · Get a FinOps review
LLM pricing · Models by maker · Cloud compute pricing · Compare models · Price barometer · FinOps guides · API and MCP server