DeepSeek V4.1 Flash

CHEAPFASTNEW

DeepSeek · deepseek/deepseek-v4.1-flash

Available

MIT-licensed 304B MoE model. Strongest cost-and-agent option in the open-weights space with 1M context.

Input / 1M tokens

৳36

Output / 1M tokens

৳144

Context length

Max output

384K

Placeholder rates. Final pricing is published per model before launch.

Suitable for

Fast
Coding
Low Cost
Long context

Example request

cURL

curl https://padmarouter.com/v1/chat/completions \
  -H "Authorization: Bearer $PADMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4.1-flash",
    "messages": [{"role": "user", "content": "Hello from Dhaka!"}]
  }'

Try in Playground Docs for this model

Performance

Throughput120 tok/s

Time to first token0.6s

StreamingSupported

Tool callingSupported

Measured from Dhaka over the last 24h of prototype traffic.

Routing

Primary routeDeepSeek direct

FallbackSecondary region

Prompt retentionNot retained