DeepSeek
DeepSeek-V4.1-Flash API
DeepSeek-V4.1-Flash, the open-weight model from DeepSeek, is live on ScaledThought's OpenAI-compatible API in the US region. Call it as deepseek-v4-1-flash with any OpenAI SDK, paid per token from prepaid credits your human approves.
Deep links into apps you have installed. Claude and Claude Code are trademarks of Anthropic; Codex is a trademark of OpenAI. ScaledThought is not affiliated with or endorsed by either.
Price per 1M tokens, US region
Input
$0.22
Cached input
$0.007
Output
$0.66
Read from the US region's API, refreshed at most every 5 minutes; last read . Prepaid credits, no subscription. Pricing
Specs
- Model id
deepseek-v4-1-flash- Publisher
- DeepSeek
- Weights released
- Parameters
- 552B total, 16B active
- Architecture
- MoE
- Context
- 1M tokens (1,000,000)
- Region
- US
- Licence
- MIT ↗
Use it
Call it with any OpenAI SDK.
Your agent sets up the account and writes ST_API_KEY and ST_BASE_URL to ./.env. Then pass the model id deepseek-v4-1-flash.
import os
from openai import OpenAI
client = OpenAI(base_url=os.environ["ST_BASE_URL"], api_key=os.environ["ST_API_KEY"])
resp = client.chat.completions.create(
model="deepseek-v4-1-flash",
messages=[{"role": "user", "content": "Hello"}],
)Licence
DeepSeek-V4.1-Flash is published by DeepSeek under the MIT. When you use it through ScaledThought, you must comply with that licence and any acceptable-use policy it includes. The publisher's licence text applies; our label is only a summary. Model licences
Not affiliated with or endorsed by Anthropic, OpenAI, Cursor or any model publisher. Claude, Claude Code, Codex, Cursor, OpenAI and model names are trademarks of their owners. Serving a model doesn't imply its publisher endorses ScaledThought.
Other live models