DeepSeek API for V4.1 Flash, V4 Pro and uncensored DeepSeek.
Outrun serves DeepSeek V4.1 Flash, V4 Pro and their uncensored and abliterated versions through one OpenAI-compatible API key. The lowest-priced is DeepSeek V4.1 Flash 1017 at $0.08 per 1M input tokens and $0.16 per 1M output tokens, 73–86% below DeepSeek's own list price. Every DeepSeek ID runs with zero data retention at Outrun.
$0.10free API credit for eligible new accounts.
- Lowest price · Provider ZDRDeepSeek V4.1 Flash 1017
deepseek-v4.1-flash-1017$0.08 in / $0.16 out - DeepSeek's Pro modelDeepSeek V4 Pro
deepseek-v4-pro$1.35 in / $4.05 out - UncensoredDeepSeek V4 Pro Uncensored
deepseek-v4-pro-uncensored$0.50 in / $0.50 out - AbliteratedDeepSeek V4.1 Flash Abliterated (2)
deepseek-v4.1-flash-abliterated-2$0.70 in / $0.70 out - V4 FlashDeepSeek V4 Flash
deepseek-v4-flash$0.45 in / $1.35 out
- Endpoint
- /v1/chat/completions
- Format
- OpenAI-compatible
- DeepSeek IDs
- 8
- Input from
- $0.08
- Data retention
- Zero at Outrun
Which DeepSeek model ID should I use?
Start with deepseek-v4.1-flash-1017. It is the lowest-priced DeepSeek ID on Outrun and also carries a Provider ZDR badge. Choose deepseek-v4-pro for DeepSeek's Pro model, and an uncensored or abliterated ID when you need a model that refuses less.
- Lowest cost per token
deepseek-v4.1-flash-1017$0.08 / $0.16 - DeepSeek's Pro model
deepseek-v4-pro$1.35 / $4.05 - Fewer refusals at a flat rate
deepseek-v4-pro-uncensored$0.50 / $0.50 - Refusals removed from the weights
deepseek-v4.1-flash-abliterated-2$0.70 / $0.70 - The earlier V4 Flash
deepseek-v4-flash$0.45 / $1.35
How much does the DeepSeek API cost on Outrun?
DeepSeek input prices on Outrun range from $0.08 to $2.10 per 1M tokens across 8 model IDs, 3 of them uncensored or abliterated. Every ID is billed per token from prepaid USD credit.
86%lower output price than DeepSeek's list
DeepSeek V4.1 Flash 1017 costs $0.08 in and $0.16 out per 1M tokens on Outrun. DeepSeek's pricing page lists its deepseek-flash model, version DeepSeek-V4.1-Flash-1017, at $0.30 in and $1.20 out at peak, and half that off-peak. List price checked .
| Model | Input | Output | Cached input |
|---|---|---|---|
| Standard | |||
DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-1017Provider ZDRDeepSeek lists its deepseek-flash model as this version | $0.08 | $0.16 | $0.02 |
DeepSeek V4.1 Flash 1017 (3)deepseek-v4.1-flash-1017-3Alternative route for 1017 | $0.20 | $1.00 | $0.01 |
DeepSeek V4.1 Flashdeepseek-v4.1-flash | $0.30 | $1.20 | $0.01 |
DeepSeek V4 Flashdeepseek-v4-flash | $0.45 | $1.35 | $0.02 |
DeepSeek V4 Prodeepseek-v4-pro | $1.35 | $4.05 | $0.05 |
| Uncensored and abliterated | |||
DeepSeek V4 Pro Uncensoreddeepseek-v4-pro-uncensoredUncensoredUncensored version of DeepSeek V4 Pro | $0.50 | $0.50 | $0.10 |
DeepSeek V4.1 Flash Abliterateddeepseek-v4.1-flash-abliteratedAbliteratedRefusals removed in the model weights, not by a prompt | $2.10 | $2.10 | $0.36 |
| Alternative routes | |||
DeepSeek V4.1 Flash Abliterated (2)deepseek-v4.1-flash-abliterated-2AbliteratedRefusals removed in the model weights, not by a prompt · alternative route | $0.70 | $0.70 | $0.12 |
Rates are shown rounded up to the cent; billing uses the exact rate. A model marked (2) or (3) is an alternative route with its own ID and price. The pricing page lists all 80 models.
How do I call the DeepSeek API through Outrun?
DeepSeek models on Outrun use the OpenAI Chat Completions format, so the official OpenAI SDKs and any OpenAI-compatible app work with them.
- 01
Get your key
Open the Telegram bot and copy your personal key. It works for DeepSeek and every other Outrun model.
- 02
Set the base URL
Use
https://api.outrun.to/v1exactly as shown, without/chat/completions. - 03
Send a DeepSeek model ID
Put the full ID, such as
deepseek-v4.1-flash-1017, in themodelfield.
curl https://api.outrun.to/v1/chat/completions \
-H "Authorization: Bearer $OUTRUN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4.1-flash-1017",
"messages": [
{"role": "user", "content": "Hello!"}
]
}'
Keep your key in an environment variable, never in client-side code.
Coding tools that take an OpenAI-compatible provider
Does Outrun keep my DeepSeek prompts?
No. Every DeepSeek ID, like every other model on Outrun, runs with zero data retention at Outrun: prompts and responses stay out of Outrun's request logs, and API calls are never saved to chat history. DeepSeek V4.1 Flash 1017 also carries a Provider ZDR badge, which adds zero retention at the model provider. Provider audit exceptions still apply, including records of generated text for up to 30 days.
DeepSeek API questions
The model docs list every DeepSeek ID with its current rate.
Do I need a DeepSeek account?
No. One Outrun key and one prepaid USD balance cover DeepSeek and every other model on Outrun. You get the key from the Telegram bot, and eligible new accounts start with $0.10 in free credit.
Does Outrun swap DeepSeek for a cheaper model?
No. Outrun sends each request to the DeepSeek model ID you choose and never reroutes it to a cheaper or different model behind the scenes. If you want the lowest price, choose deepseek-v4.1-flash-1017 yourself.
Can I try DeepSeek for free?
Eligible new accounts start with $0.10 in free credit, enough for about 410 requests of 2,000 input and 500 output tokens on DeepSeek V4.1 Flash 1017. The Playground also gives every account 10 free web searches a day.
Does Outrun charge less at off-peak hours?
No. Outrun bills each DeepSeek ID at the one rate shown on this page at any time of day. DeepSeek's off-peak discount applies only on DeepSeek's own API.