How much does the DeepSeek V4.1 Flash API cost?
On OceanToken, DeepSeek V4.1 Flash costs $0.21 per 1M input tokens and $0.84 per 1M output tokens at Tier 1 (30% off the official price) and $0.165 per 1M input tokens and $0.66 per 1M output tokens at Tier 3: 45% off DeepSeek’s official price of $0.30 input / $1.20 output. Cached input bills at the official $0.006 per 1M tokens at every tier. One request with 100K input and 20K output tokens costs $0.0378 at Tier 1 and $0.0297 at Tier 3, against $0.054 at the official price.
| per 1M tokens | Official price | Tier 1 | Tier 2 | Tier 3 |
|---|---|---|---|---|
| Input | $0.30 | $0.21 −30% | $0.195 −35% | $0.165 −45% |
| Output | $1.20 | $0.84 −30% | $0.78 −35% | $0.66 −45% |
| Cached inputOfficial price at every tier | $0.006 | $0.006 | $0.006 | $0.006 |
Prices as of November 18, 2026.
What does a typical request cost?
Tier 1 is below $1,000 in cumulative top-ups, Tier 2 starts at $1,000 and Tier 3 at $10,000. Gifts and credit grants count toward the total. Your account moves up on its own, and every later request bills at the lower rate.
| Workload | Official price | Tier 1 | Tier 2 | Tier 3 |
|---|---|---|---|---|
| One request | $0.054 | $0.0378 −30% | $0.0351 −35% | $0.0297 −45% |
| 1,000 requests | $54.00 | $37.80 −30% | $35.10 −35% | $29.70 −45% |
How to pay the least for DeepSeek V4.1 Flash
Three things set what a DeepSeek V4.1 Flash request costs on OceanToken: your tier, how much the model thinks, and how much of the input comes from the cache. Each effect below is worked out from the prices on this page.
| Lever | What to do | Effect on the bill |
|---|---|---|
| Your tier | Tier 3 starts at $10,000 in cumulative top-ups; gifts and credit grants count toward it, and the account moves up on its own. | $0.165 input / $0.66 output per 1M, −45% against DeepSeek's $0.30 / $1.20 |
| Less thinking | Send reasoning_effort: "low" and leave room in max_tokens: reasoning tokens bill as output. | −36% per answer in our own test of November 14, 2026 (2,452 output tokens instead of 3,858) |
| Cached input | Repeated input read from the cache bills at DeepSeek's own cache price, the same at every tier. | $0.006 per 1M, −97% against Tier 1 input at $0.21 |
The thinking test in full, with code
How fast is DeepSeek V4.1 Flash? Measured on our API
Measured on the OceanToken API on November 16, 2026: the median of 3 streaming runs per model, one short prompt, up to 150 output tokens. Longer prompts and outputs take longer.
| Model | Time to first token | Output speed |
|---|---|---|
| deepseek-v4.1-flash | 0.93 s | 108 tokens/s |
| DeepSeek V4 Pro | 1.59 s | 52 tokens/s |
Source: OceanToken /speed, measured November 16, 2026. Every measured model
Large volume? Enterprise rates are by quote, up to 60% off. Quote by email
Frequently asked questions
How much does the DeepSeek V4.1 Flash API cost per million tokens, and where can I get it cheaper than the official price?
DeepSeek V4.1 Flash costs $0.30 per 1M input tokens and $1.20 per 1M output tokens at DeepSeek’s official price. On OceanToken it costs $0.21 input and $0.84 output at Tier 1 (30% off) and $0.165 / $0.66 at Tier 3 (45% off). Cached input bills at the official $0.006 per 1M tokens. You call the same model with one OpenAI-compatible API key.
DeepSeek V4.1 Flash vs DeepSeek V4 Pro: which should I use?
At Tier 1 on OceanToken, DeepSeek V4.1 Flash costs $0.21 input / $0.84 output per 1M tokens and DeepSeek V4 Pro costs $1.056 / $3.168. A request with 100K input and 20K output tokens costs $0.0378 on DeepSeek V4.1 Flash and $0.169 on DeepSeek V4 Pro. Start with the one that costs less and switch only if the other does better on your own tests; both take the same API key, so switching is a change of model ID.
Is cached input discounted on DeepSeek V4.1 Flash?
No. Cached input bills at DeepSeek’s official price at every tier: $0.006 per 1M tokens for DeepSeek V4.1 Flash. The discounts apply to input and output.
How do I get the Tier 3 price?
Top up more. Tier 1 is below $1,000 in cumulative top-ups, Tier 2 starts at $1,000 and Tier 3 at $10,000. Gifts and credit grants count toward the total. Your account moves up on its own, and the console shows the rate that applies to you.
Am I charged for failed requests?
No. If a request fails, you are not charged: that covers a request refused for an insufficient balance or by a safety check, one whose provider fails before returning any output, and a video job that fails or is refused by a safety check, whose held amount goes back to your balance. The one exception: if a request breaks off after the model has started producing output, for example a stream cut off part-way, you pay for the tokens that were processed and generated for it.
Is there a volume discount?
Yes. Enterprise rates are by quote, up to 60% off: ask for a quote by email on the Enterprise page and we activate your rate for you.
Is it the official DeepSeek V4.1 Flash model?
Yes. You call DeepSeek’s DeepSeek V4.1 Flash by its model ID, deepseek/deepseek-v4.1-flash, and get that model. The price table shows the official price next to ours.
Do I need a subscription to use DeepSeek V4.1 Flash?
No. You add credits by card, and each request is billed at the model’s price. Purchased credits are valid for 12 months.
How do I call the DeepSeek V4.1 Flash API?
Create an API key in the OceanToken console, set the base URL to https://api.oceantoken.ai/v1 and use the model id, for example deepseek/deepseek-v4.1-flash. The OpenAI SDKs and any OpenAI-compatible client work.
from openai import OpenAI
# Your API Key from the OceanToken console
client = OpenAI(base_url="https://api.oceantoken.ai/v1",
api_key="YOUR_API_KEY")
reply = client.chat.completions.create(
model="deepseek/deepseek-v4.1-flash",
messages=[{"role": "user", "content": "Hello"}])
print(reply.choices[0].message.content)# Your API Key from the OceanToken console
curl https://api.oceantoken.ai/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "deepseek/deepseek-v4.1-flash",
"messages": [{"role": "user", "content": "Hello"}]}'Read next
- GPT Empty Response? Fix max_output_tokens
- DeepSeek API 429 Rate Limit and Server Busy: How to Fix
- Limit LLM API Spend With a Prepaid Balance
DeepSeek API: every DeepSeek model and tier
Keep exploring
DeepSeek V4.1 Flash model page · DeepSeek API · All model prices
DeepSeek V4.1 Flash vs Gemini 3.5 Flash · DeepSeek V4.1 Flash vs Qwen3.8 Flash · DeepSeek V4.1 Flash vs MiniMax M3 · DeepSeek V4.1 Flash vs Claude Opus 5.5 · DeepSeek V4.1 Flash vs Claude Fable 5.2 · DeepSeek V4.1 Flash vs Claude Fable 5.1 · All comparisons
Claude Sonnet 5.5 API price · Claude Opus 5.5 API price · GPT-5.5 API price · GPT-6 Sol API price · Gemini 3.1 Pro API price · Seedance 2.5 API price · MiniMax H3 API price · Kling v3.0 API price