DeepSeek · Text
DeepSeek V4.1 Flash API pricing
DeepSeek V4.1 Flash through one prepaid, OpenAI-compatible key: $0.15 input, $0.60 output per 1M tokens, the provider’s list price with nothing added per request.
DeepSeek V4.1 Flash price per 1M tokens
- Input
- $0.15
- per 1M tokens
- Cached input
- $0.003
- per 1M tokens, on a cache hit
- Output
- $0.60
- per 1M tokens
2× at peak DeepSeek charges double Mon–Fri 01:00–04:00 and 06:00–10:00 UTC. You pay the rate in force when the request runs.
| Per 1M tokens | Off-peak | Peak |
|---|---|---|
| Input | $0.15 | $0.30 |
| Cached input | $0.003 | $0.006 |
| Output | $0.60 | $1.20 |
DeepSeek’s list price, in US dollars. Tokens.Pay adds a 5% fee (minimum $0.79) when you add credit, never per request.
What DeepSeek V4.1 Flash is good at
The provider's recommended tier. 1M context, image understanding, tool calls.
In short: fast everyday reasoning, 1M context.
- Tool calls
- JSON output
- Image understanding
- Thinking mode
Text model, called at /v1/chat/completions. Listed under text generation, reasoning and vision.
DeepSeek V4.1 Flash context window and limits
- Provider
- DeepSeek
- Type
- Text
- Context window
- 1M tokens
- Max output
- 384K tokens
- Endpoint
/v1/chat/completions- Model id
deepseek-flash
What DeepSeek V4.1 Flash costs per request
Typical requests at the list rate, worked out once. Your own counts go in the calculator below.
| Request | Input | Of which cached | Output | Per request | Per 1,000 |
|---|---|---|---|---|---|
| Short chat reply | 1,500 | — | 300 | $0.000405 | $0.4050 |
| Support bot, cached system prompt | 6,000 | 5,000 | 500 | $0.000465 | $0.4650 |
| Summarise a 100-page document | 60,000 | — | 1,500 | $0.009900 | $9.90 |
| Agent step, long cached context | 200,000 | 180,000 | 2,000 | $0.004740 | $4.74 |
Token counts are examples; the rates are DeepSeek’s list prices off-peak (peak hours cost double). Tokens.Pay adds nothing per request.
DeepSeek V4.1 Flash cost calculator
Set the size of a typical request. Prices are per token, so the estimate is exact for those counts.
Call DeepSeek V4.1 Flash with the OpenAI SDK
Base URL https://api.airuntime.dev/v1, your key as the bearer token, and model set to deepseek-flash.
curl
curl https://api.airuntime.dev/v1/chat/completions \
-H "Authorization: Bearer $TOKENS_PAY_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-flash",
"messages": [{"role": "user", "content": "Say hello in five words."}]
}'
Python
from openai import OpenAI
import os
client = OpenAI(base_url="https://api.airuntime.dev/v1", api_key=os.environ["TOKENS_PAY_KEY"])
reply = client.chat.completions.create(
model="deepseek-flash",
messages=[{"role": "user", "content": "Say hello in five words."}],
)
print(reply.choices[0].message.content)
Node
import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://api.airuntime.dev/v1", apiKey: process.env.TOKENS_PAY_KEY });
const reply = await client.chat.completions.create({
model: "deepseek-flash",
messages: [{ role: "user", content: "Say hello in five words." }],
});
console.log(reply.choices[0].message.content);
Who processes DeepSeek V4.1 Flash requests
Your code calls AI Runtime directly. It forwards the request to the provider and keeps only the bill.
Processing and storage
- Model provider
- DeepSeek, mainland China · data terms
- Fallback
- Fireworks AI, United States · data terms
- Fallback
- DeepInfra, United States · data terms
- Gateway
- AI Runtime, AWS us-east-1
DeepSeek’s own API serves the model first. If it cannot, AI Runtime sends the request to Fireworks AI or DeepInfra, which run the same open model in the US.
Not stored by Tokens.Pay or AI Runtime. Prompts and outputs exist in memory only while the request runs; we keep the billing record. The provider processes the request under its own terms. Zero data retention, exactly
Compare DeepSeek V4.1 Flash with similar models
| Model | Provider | Input /1M | Cached /1M | Output /1M | Context |
|---|---|---|---|---|---|
| DeepSeek V4.1 Flash (this page) | DeepSeek | $0.15 | $0.003 | $0.60 | 1M |
| GPT-6 Luna | OpenAI | $0.10 | $0.01 | $0.50 | 1.05M |
| GPT-6.1 Terra | OpenAI | $1.00 | $0.05 | $5.00 | 1.05M |
| GPT-6.1 Sol | OpenAI | $2.00 | $0.10 | $10.00 | 1.05M |
| Kimi K3.1 Pro | Moonshot AI | $1.92 | $0.192 | $10.04 | 1M |
| Kimi K3.1 Pro Highspeed | Moonshot AI | $3.84 | $0.192 | $20.08 | 1M |
DeepSeek V4.1 Flash API questions
How much does the DeepSeek V4.1 Flash API cost?
$0.15 per 1M input tokens, $0.003 per 1M cached input tokens and $0.60 per 1M output tokens: DeepSeek’s list price off-peak (it doubles at the provider’s weekday peak). Tokens.Pay adds nothing per request; a 5% fee (minimum $0.79) applies when you add credit.
What is the model id for DeepSeek V4.1 Flash?
Send model "deepseek-flash" to https://api.airuntime.dev/v1/chat/completions with your Tokens.Pay key as the bearer token. Any OpenAI-compatible SDK works.
What is the DeepSeek V4.1 Flash context window?
1M tokens, with up to 384K output tokens per request.
Does Tokens.Pay store my DeepSeek V4.1 Flash prompts?
No. AI Runtime, the gateway your key calls, keeps prompts and outputs in memory only while the request runs and stores just the billing record. The request is processed by DeepSeek (mainland China), or as a fallback Fireworks AI or DeepInfra (United States), under the provider’s own API data terms.
Does DeepSeek V4.1 Flash support prompt caching?
Yes. Input tokens served from DeepSeek’s cache are billed at $0.003 per 1M instead of $0.15. Your Dashboard shows your cache hit rate.
Add credit, then call DeepSeek V4.1 Flash.
From $5, plus a 5% top-up fee (minimum $0.79). No subscription; paid credit never expires.
Add credit