---
title: "DeepSeek V4.1 Flash API pricing, per token and plans"
description: "DeepSeek V4.1 Flash API pricing per 1M input, cached and output tokens, flat monthly, weekly and hourly plans, and how DeepSeek, DeepInfra and Fireworks compare."
url: https://broadrouter.com/blog/deepseek-v4-1-flash-api-pricing/
lang: en
updated: 2026-11-14
publisher: BroadRouter (a BroadNet site); API by BroadNet Inference, https://inference.broadnet.ai
---

# DeepSeek V4.1 Flash API pricing: per token, cached and plans

By the BroadNet Inference team · 5 min read · Published 2026-11-14 · Updated 2026-11-16

On BroadNet Inference, DeepSeek V4.1 Flash costs $0.175 per 1M input tokens, $0.0055 per 1M cached input tokens and $0.52 per 1M output tokens, reasoning included. Or pay a flat price per session: $60 a month, $15 a week or $1 an hour, with no token meter, one request at a time.

## Key takeaways

-   Pay per token: $0.175 input, $0.0055 cached input, $0.52 output, per 1M tokens.
-   Flat plans: Full Moon $60 a month, Orbit $15 a week, Meteor $1 an hour, per session, unlimited tokens.
-   Context is 1,048,576 tokens, with text and image input. Reasoning is billed as output on per-token keys.
-   A full-time coding agent in our example costs $65.40 a month per token, so Full Moon wins; a side project costs $14.23, so per token wins.
-   DeepSeek V4.1 Flash needs a top-up; the $5 starting credit covers other models.

**On this page**

1.  How much does DeepSeek V4.1 Flash cost per token?
2.  What do the flat plans cost?
3.  What does one request cost?
4.  What does a month cost per token?
5.  How much can one plan session process?
6.  How does this compare with other providers?
7.  What else affects the bill?
8.  Which DeepSeek V4.1 Flash pricing option should you choose?
9.  Where BroadNet Inference fits

DeepSeek V4.1 Flash is the model on BroadNet’s unlimited plans and one of four chat models sold per token. That gives you four ways to pay for the same model: per token, or a monthly, weekly or hourly plan. This page lists all of them, works through real requests, and shows where each option wins. Prices come from BroadNet’s [pricing file](https://inference.broadnet.ai/pricing.json?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-deepseek-v4-1-flash-api-pricing&utm_content=body), synced 2026-11-14. For the same model’s price at every host we track, see the [DeepSeek V4.1 Flash price page](https://broadrouter.com/llm-api-pricing/deepseek-v4.1-flash/). For how it sits against the rest of the market, see the [LLM API price index](https://broadrouter.com/data/llm-api-price-index/).

## How much does DeepSeek V4.1 Flash cost per token?

| Token type | Price per 1M tokens |
| --- | --- |
| Input | $0.175 |
| [Cached input](https://broadrouter.com/glossary/cached-input-tokens/) | $0.0055 |
| Output, including [reasoning](https://broadrouter.com/glossary/reasoning-tokens/) | $0.52 |

Model ID: `broadnet/deepseek-v4.1-flash`. Context: 1,048,576 tokens per request, prompt and answer together. Input: text and images, with images billed as input tokens. A pay-as-you-go key allows 30 requests a minute and 4 in flight.

## What do the flat plans cost?

| Plan | Price per session | Period | Paid by |
| --- | --- | --- | --- |
| Full Moon | $60 | Month, renews | Card |
| Orbit | $15 | Week | Account balance |
| Meteor | $1 | One hour from purchase | Account balance |

Each session gives unlimited DeepSeek V4.1 Flash tokens, one request at a time. There is no token meter, no monthly cap, no 5-hour window and no weekly limit. With a founders code, Full Moon is $45 a month for the first three months. Running Orbit every week for a month works out to $64.29, so Full Moon is the cheaper way to cover a whole month. [How sessions work](https://broadrouter.com/guides/unlimited-llm-api-sessions/).

## What does one request cost?

Three typical shapes, sized to land on round token totals:

| Workload | Tokens | Cost |
| --- | --- | --- |
| 100 chat requests of 10,000 input and 1,000 output tokens | 1M input, 0.1M output | $0.175 plus a tenth of $0.52 |
| 10 agent turns re-sending a 100,000-token cached repository | 1M cached input | $0.0055 |
| 1,000 answers of 1,000 tokens, reasoning included | 1M output | $0.52 |

The cache line is the one that matters for agents. Cached input costs $0.0055 per 1M on this model, against $0.175 for new input, so a stable prompt prefix is usually the biggest single saving. [How to raise your cache rate](https://broadrouter.com/blog/cached-input-tokens-explained/), or model your own hit rate in the [prompt caching calculator](https://broadrouter.com/tools/prompt-caching-calculator/).

## What does a month cost per token?

BroadNet’s worked scenarios, priced on the per-token rates above:

| Workload | Input a month | Cache hits | Output a month | Per token | Cheaper option |
| --- | --- | --- | --- | --- | --- |
| Side project | 100M | 50% | 10M | $14.23 | pay as you go |
| Daily assistant | 300M | 50% | 30M | $42.67 | pay as you go |
| Full-time coding agent | 1,000M | 80% | 50M | $65.40 | Full Moon |
| Always-on agent | 3,000M | 80% | 150M | $196.20 | Full Moon |

At 80% cache hits and 20 input tokens per output token, one Full Moon session breaks even at about 917 million input and 46 million output tokens a month. [The full break-even math](https://broadrouter.com/blog/flat-rate-vs-pay-per-token-llm-api/), or plug in your own volumes with the [flat-rate vs per-token calculator](https://broadrouter.com/tools/flat-rate-vs-per-token-calculator/).

## How much can one plan session process?

A session runs one request at a time. At a typical speed of about 100 tokens per second, a session generating non-stop for 30 days would produce about 259 million output tokens. Real agents read, run tools and wait, so the practical figure is lower. To go faster, run more sessions in parallel; [parallel agents on a flat rate](https://broadrouter.com/blog/run-parallel-coding-agents-flat-rate/) explains how many you need.

## How does this compare with other providers?

Published prices for the same model, per 1M tokens, fetched 14 November 2026 from each provider’s pricing page. DeepSeek’s own API now takes the model name `deepseek-flash` (version DeepSeek-V4.1-Flash-1017). DeepInfra shows its rates as discounts from a list price of USD 0.20 / 0.006 / 0.60 and does not say when the discounts end:

| Provider | Input | Cached input | Output |
| --- | --- | --- | --- |
| BroadNet Inference | $0.175 | $0.0055 | $0.52 |
| DeepSeek API, peak hours | USD 0.30 | USD 0.006 | USD 1.20 |
| DeepSeek API, off-peak | USD 0.15 | USD 0.003 | USD 0.60 |
| Together AI | USD 0.30 | USD 0.006 | USD 1.20 |
| Fireworks AI, standard | USD 0.30 | USD 0.006 | USD 1.20 |
| Fireworks AI, spot | USD 0.21 | USD 0.0042 | USD 0.84 |
| DeepInfra, DeepSeek-V4.1-Flash, 30% off list | USD 0.14 | USD 0.0042 | USD 0.42 |
| DeepInfra, DeepSeek-V4.1-Flash-1017, 51% off list | USD 0.098 | USD 0.0029 | USD 0.294 |

BroadNet’s rates are below Together’s, Fireworks’ and DeepSeek’s peak rates, but not the lowest. DeepInfra’s discounted listings are cheaper than BroadNet on input, cached input and output at any hour. DeepSeek’s own API halves its prices outside its peak windows (01:00 to 04:00 and 06:00 to 10:00 UTC on weekdays), which makes its off-peak input and cached input cheaper than BroadNet’s, while its output stays higher. Fireworks’ spot tier is also cheaper on cached input. Routers list more offers; [the provider pricing comparison](https://broadrouter.com/blog/llm-inference-provider-pricing-compared/) covers them with sources.

## What else affects the bill?

-   **Reasoning.** DeepSeek V4.1 Flash thinks before answering. Set `reasoning_effort` to `low` for simple steps, and keep `max_tokens` at 1,000 or more. [Why replies come back empty](https://broadrouter.com/guides/reasoning-tokens-empty-response/).
-   **Retries.** A retried request is a new inference and is billed again on per-token keys.
-   **Images.** Billed as input tokens on per-token keys, included on plans.
-   **Long context.** Every turn resends the context. Caching keeps that cheap; [when a 1M context is worth it](https://broadrouter.com/blog/million-token-context-window-guide/).

## Which DeepSeek V4.1 Flash pricing option should you choose?

1.  Start on pay-as-you-go and log a week of `prompt_tokens`, `cached_tokens` and `completion_tokens`.
2.  Multiply by about 4.3 and put the totals in the [LLM API cost calculator](https://broadrouter.com/tools/llm-api-cost-calculator/) or the [BroadNet pricing page](https://broadrouter.com/pricing/).
3.  If a plan wins, test your agent on a $1 Meteor hour. [The Aider setup guide](https://broadrouter.com/guides/aider-unlimited-deepseek/) gets a terminal agent running in minutes.
4.  If you need tools and structured output listed by BroadNet, also compare [GLM 5.3 vs DeepSeek V4.1 Flash](https://broadrouter.com/blog/glm-5-3-vs-deepseek-v4-1-flash/). For other ways to reach DeepSeek models, see [DeepSeek API alternatives](https://broadrouter.com/blog/deepseek-api-alternatives/).

## Where BroadNet Inference fits

BroadNet is the option if you want DeepSeek V4.1 Flash both per token and unlimited on one account, with a 1M context and [zero data retention](https://broadrouter.com/zero-data-retention/). It is not the cheapest per-token price: DeepInfra’s discounted listings are lower on every rate, and DeepSeek’s off-peak input and cached rates are lower. For light or bursty traffic billed per token, those may cost less. BroadNet’s edge is the flat plan once an agent runs most of the day, plus no content retained. [Choose a plan](https://inference.broadnet.ai/plans?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-deepseek-v4-1-flash-api-pricing&utm_content=body).

Free tools for this

-   [LLM API cost calculator](https://broadrouter.com/tools/llm-api-cost-calculator/) Price a workload on every model at once, with cache hits, reasoning and a flat-plan check.
-   [Flat-rate vs per-token LLM cost calculator](https://broadrouter.com/tools/flat-rate-vs-per-token-calculator/) Break-even between an unlimited plan and paying per token, per agent.
-   [Prompt caching savings calculator](https://broadrouter.com/tools/prompt-caching-calculator/) Bill with and without caching, by prefix share and hit rate.
-   [Prompt token diff](https://broadrouter.com/tools/prompt-token-diff/) Two prompt versions side by side: changed lines, exact token delta and the monthly cost difference.

[All free tools](https://broadrouter.com/tools/)

## Terms in this article

[Token](https://broadrouter.com/glossary/token/)

A token is the unit of text a language model reads and writes: a word, part of a word, a punctuation mark or a byte sequence produced by the model's tokenizer. LLM APIs count, limit and usually bill usage in tokens.

[Pay-as-you-go](https://broadrouter.com/glossary/pay-as-you-go/)

Pay-as-you-go is a pricing model in which you pay only for what you use, metered per unit such as tokens, requests or audio minutes, with no subscription or minimum commitment. Most LLM APIs bill this way, from a prepaid balance or a periodic invoice.

[Input and output tokens](https://broadrouter.com/glossary/input-and-output-tokens/)

Input tokens are the tokens a language model reads in a request, including the system prompt, conversation history, tool definitions and images. Output tokens are the tokens it generates in reply, including any reasoning. LLM APIs usually price them separately, and output usually costs more.

[Tool calling](https://broadrouter.com/glossary/tool-calling/)

Tool calling, also called function calling, is an LLM API feature in which the model returns a structured request to run a function the developer has defined, with JSON arguments. The application runs the function and sends the result back; the model never runs code itself.

[Cached input tokens](https://broadrouter.com/glossary/cached-input-tokens/)

Cached input tokens are prompt tokens that an LLM provider has recently processed and can reuse from a cache instead of computing again, usually a repeated prefix such as a system prompt or document. Providers bill them at a lower rate than new input tokens.

[Prepaid credit](https://broadrouter.com/glossary/prepaid-credit/)

Prepaid credit is money added to an API account before use. Each request deducts its cost from the balance, and when the balance runs out, metered requests stop until you top up. It caps spend at what you have paid in.

[AI coding agent](https://broadrouter.com/glossary/coding-agent/)

An AI coding agent is a software tool in which a language model works on a codebase in a loop: it reads files, edits them, runs commands such as tests, reads the output and decides the next step until the task is done or a limit is reached.

[Concurrency](https://broadrouter.com/glossary/concurrency/)

Concurrency is the number of requests a client has in flight at the same time. LLM APIs often cap it per key or plan, separately from requests per minute, because each open request holds GPU capacity until its response has finished.

## Questions

### How much does DeepSeek V4.1 Flash cost per million tokens?

On BroadNet Inference, $0.175 per 1M input tokens, $0.0055 per 1M cached input tokens and $0.52 per 1M output tokens.

### Is there an unlimited DeepSeek API plan?

Yes. BroadNet's Full Moon plan is $60 a month per session with no token meter, no monthly cap and no weekly limit. One session runs one request at a time.

### Are reasoning tokens extra?

On per-token keys, reasoning is reported in completion tokens and billed at the output price. On a plan, there is no token meter, so reasoning costs nothing extra.

### Can I use DeepSeek V4.1 Flash on the free credit?

No. DeepSeek V4.1 Flash is available after a top-up. The $5 starting credit covers Gemma 4 VL 26B, the two transcription models and hourly keys.

### When are DeepSeek API off-peak hours?

DeepSeek's pricing page (fetched 14 November 2026) lists peak hours as 01:00 to 04:00 and 06:00 to 10:00 UTC, Monday to Friday, excluding Chinese public holidays. All other hours, including weekends, are billed at half the peak rate.

### How does BroadNet's price compare with DeepSeek's own API?

BroadNet is not the cheapest per token. DeepSeek's off-peak input and cached input are cheaper than BroadNet's (its output is dearer), and DeepInfra lists a discounted DeepSeek-V4.1-Flash-1017 below BroadNet on all three rates. BroadNet's edge is the flat plan at high volume, zero data retention and one key for chat and speech to text.

## Keep reading

-   [Flat-rate vs pay-per-token LLM API: the break-even math](https://broadrouter.com/blog/flat-rate-vs-pay-per-token-llm-api/)
-   [Cached input tokens: how to cut an LLM API bill](https://broadrouter.com/blog/cached-input-tokens-explained/)
-   [GLM 5.3 vs DeepSeek V4.1 Flash: which one to use, and when](https://broadrouter.com/blog/glm-5-3-vs-deepseek-v4-1-flash/)
-   [How inference providers price open models: 6 compared](https://broadrouter.com/blog/llm-inference-provider-pricing-compared/)
-   [LLM API price changes in October 2026: 3 rises, no cuts](https://broadrouter.com/blog/llm-api-price-changes-october-2026/)
-   [Do LLM API prices only go down? 52 rises and how to budget for them](https://broadrouter.com/blog/llm-api-price-increases/)
-   [Guide: How the unlimited LLM API plans work for agents](https://broadrouter.com/guides/unlimited-llm-api-sessions/)
-   [Guide: Run Aider on unlimited DeepSeek V4.1 Flash](https://broadrouter.com/guides/aider-unlimited-deepseek/)
-   [Guide: Why a reasoning model returns an empty answer](https://broadrouter.com/guides/reasoning-tokens-empty-response/)

Try it on your real workload. New accounts start with $5 of credit for Gemma and transcription. DeepSeek, GLM and Qwen3.8 need a top-up.

[Get started](https://inference.broadnet.ai/account/?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-deepseek-v4-1-flash-api-pricing&utm_content=page-cta) ·

**Sources**

-   [BroadNet Inference machine-readable prices (pricing.json, fetched 2026-11-14)](https://inference.broadnet.ai/pricing.json?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-deepseek-v4-1-flash-api-pricing&utm_content=sources)
-   [BroadNet Inference plans and plan FAQ (fetched 2026-11-14)](https://inference.broadnet.ai/plans?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-deepseek-v4-1-flash-api-pricing&utm_content=sources)
-   [BroadNet Inference API reference, limits (fetched 2026-11-14)](https://inference.broadnet.ai/docs?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-deepseek-v4-1-flash-api-pricing&utm_content=sources#limits)
-   [DeepSeek API docs, models and pricing, peak and off-peak (fetched 2026-11-14)](https://api-docs.deepseek.com/quick_start/pricing)
-   [Together AI serverless pricing (fetched 2026-11-14)](https://www.together.ai/pricing)
-   [Fireworks AI pricing, standard and spot (fetched 2026-11-14)](https://fireworks.ai/pricing)
-   [DeepInfra pricing, discounted DeepSeek listings (fetched 2026-11-14)](https://deepinfra.com/pricing)
