Models · DeepSeek
DeepSeek V4.1 Flash API, 1.8x less than the official price
DeepSeek V4.1 Flash costs $0.165 per million input tokens and $0.66 per million output tokens on Kontinent AI. DeepSeek's official price is $0.30 and $1.20, so you pay 1.8x less (45% less).
- No subscription
- Pay only for requests that get an answer
- Your balance does not expire
- Sign in with Google or email
Price per million tokens
| Tokens | Official price | Kontinent AI |
|---|---|---|
| Input | Official price | $0.165 |
| Output | Official price | $0.66 |
| Cached input | Official price | $0.0033 |
USD per million tokens. The official price is DeepSeek's, on its price page. Ours is that price times 0.55, for every kind of token. This model thinks before it answers. The thinking is billed as output tokens, at the same multiple.
What jobs cost with DeepSeek V4.1 Flash
| Job | Official price, then ours |
|---|---|
| 1,000 support-bot replies 1,500 tokens in and 500 out, each | Official price |
| One coding-agent session 200,000 tokens in and 20,000 out, uncached | Official price |
| 1,000 document summaries 4,000 tokens in and 300 out, each | Official price |
Each job is priced the way the API bills it. Work out your monthly bill.
Run DeepSeek V4.1 Flash for $0.05 a session, not $0.08
The same DeepSeek V4.1 Flash, through one API key. Change the base URL and the key, and keep your code. You see what every request cost and what you saved.
- No subscription
- Pay only for requests that get an answer
- Your balance does not expire
- Sign in with Google or email
Your monthly bill
Enter what you spend at the official price and see it on Kontinent AI, in your currency.
1.8x less than the official price
- At the official price
- $100.00
- With Kontinent AI
- $55.00
- You keep
- $45.00
The model
- Maker
- DeepSeek
- Model id
- deepseek-v4.1-flash
- Context window
- 1,000,000 tokens
- Weights
- Open weights
- Beyond plain chat
- Tool calling
- Speed, measured by us
- 2,951 ms to first token · 38.1 tokens/s (19 Oct)
Use DeepSeek V4.1 Flash
POST /v1/chat/completions: the OpenAI SDKs, OpenCode, most OpenAI-compatible tools.POST /v1/messages: Claude Code, the Anthropic SDKs.
curl https://api.kontinentai.com/v1/chat/completions \
-H "Authorization: Bearer $KONTINENT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "deepseek-v4.1-flash", "messages": [{"role": "user", "content": "Hello"}]}'export ANTHROPIC_BASE_URL=https://api.kontinentai.com
export ANTHROPIC_AUTH_TOKEN=$KONTINENT_API_KEY
claude --model deepseek-v4.1-flashSetup for each tool: Claude Code, Codex, OpenCode, OpenAI SDK, Anthropic SDK.
Questions
How much does the DeepSeek V4.1 Flash API cost?
DeepSeek V4.1 Flash costs $0.165 per million input tokens and $0.66 per million output tokens on Kontinent AI. DeepSeek's official price is $0.30 and $1.20, so you pay 1.8x less (45% less). Cached input is $0.0033 per million tokens, against $0.006 official.
Is it the same DeepSeek V4.1 Flash?
Yes. We send your request to the model you name, and nothing else. We check the name of the model that answered on each response.
What does a coding-agent session cost with DeepSeek V4.1 Flash?
A session of 200,000 input tokens and 20,000 output tokens, uncached, costs $0.05 on Kontinent AI and $0.08 at DeepSeek's official price.
Can I use DeepSeek V4.1 Flash with OpenCode or Claude Code?
Yes. Point OpenCode, Claude Code at Kontinent AI by changing the base URL and the key. The setup pages in the docs show each one.
What is the context window of DeepSeek V4.1 Flash?
1M tokens (1,000,000), as DeepSeek publishes it.
Compare DeepSeek V4.1 Flash
Other models
Start with DeepSeek V4.1 Flash today
DeepSeek V4.1 Flash is 45% less than DeepSeek's official price on every token. Top up a prepaid balance and call it from the tools you already use.
- No subscription
- Pay only for requests that get an answer
- Your balance does not expire
- Sign in with Google or email