DeepSeek Budget Operational · 100% success (24h)

DeepSeek V4 Flash 0731 API: pricing, code samples & live stats

DeepSeek V4 Flash 0731 from the DeepSeek family, Budget class. Call it with one BangunWeb API key in OpenAI or Anthropic format. No credit card needed, and usage is billed in prepaid tokens.

Model ID

deepseek-v4-flash-0731

Sign up with Google, no payment. API key ready instantly.

Price per 1M tokens

Input tokens$0.031
Output tokens$0.043
Base fee per request
$0.0001
Example request (1,000 in + 500 out)
$0.0001
One coding-agent step (~20k input tokens)
$0.0007

Based on the Saver package. Bigger packages cost less: from $0.018 per 1M input tokens on Raja.

Packages from $0.55. Pay with QRIS, Indonesian bank transfer, or crypto (USDT). Tokens never expire.

Top Model rank

Real usage, last 7 days

From customer requests through the BangunWeb gateway, not vendor numbers. Refreshed every 15 minutes.

#16Top Model rank
9Active users
2,313Requests
99.4%Success
29.2 sAvg response time

Time to first reply (fastest route probe): 5.9 s. See the Top Model board →

API

Use DeepSeek V4 Flash 0731 in one minute

Base URL https://router.bangunweb.com/v1, an API key from your dashboard, and model ID deepseek-v4-flash-0731. Copy, swap the key, run.

curl
curl https://router.bangunweb.com/v1/chat/completions \
  -H "Authorization: Bearer bwr_live_..." \
  -H "Content-Type: application/json" \
  -d '{"model": "deepseek-v4-flash-0731", "messages": [{"role": "user", "content": "Hi, introduce yourself briefly."}]}'
Python (OpenAI SDK)
from openai import OpenAI

client = OpenAI(
    base_url="https://router.bangunweb.com/v1",
    api_key="bwr_live_...",
)

r = client.chat.completions.create(
    model="deepseek-v4-flash-0731",
    messages=[{"role": "user", "content": "Hi, introduce yourself briefly."}],
)
print(r.choices[0].message.content)

One line installs OpenCode (terminal coding agent) with this model as default:

OpenCode
curl -fsSL https://bangunweb.com/install | BANGUNWEB_MODEL=deepseek-v4-flash-0731 sh

API docs →Beginner guide →

Similar models

One API key for all of them. Just change the model ID.

See all 65 models →All DeepSeek models and prices →

Questions about the DeepSeek V4 Flash 0731 API

How much does the DeepSeek V4 Flash 0731 API cost?

$0.031 per 1M input tokens and $0.043 per 1M output tokens (Saver package), plus a $0.0001 base fee per request. Bigger packages are cheaper per token.

How do I pay?

Buy a token package in the dashboard, from $0.55. Pay with QRIS, Indonesian bank transfer, or crypto (USDT). No credit card needed.

Does DeepSeek V4 Flash 0731 work with the OpenAI SDK?

Yes. Set base_url to https://router.bangunweb.com/v1, use your BangunWeb API key and model="deepseek-v4-flash-0731". The Anthropic format (/v1/messages) is supported too.

Can I use it in Claude Code or OpenCode?

Yes. Run the one-line installer above, or set the base URL and API key in the tool yourself. Cursor, Cline, n8n and other OpenAI-compatible tools work too.

Is there a free trial?

Sign up with a Google account to get 2M free trial tokens, enough to test this model before buying.

Try it free today

2M free tokens for new Google accounts. Start telling an AI to build your app right away.