DeepSeek V4.1 Flash

DeepSeek V4.1 Flash (deepseek/deepseek-v4.1-flash) — a light DeepSeek V4.1 tier with the family’s cheapest input: it reasons (low/high/max) and, unlike the senior models, reasoning can be fully turned off (off). Accepts images; files — no.

Parameters and capabilities

Parameter	Value
model	deepseek/deepseek-v4.1-flash (short alias deepseek-v4.1-flash)
reasoning_effort	off, low, high, max
Disabling reasoning	yes — off saves tokens on simple tasks
Images on input	yes (image_url, data URLs)
Files on input	no — a file part returns 400
Cache read	$0.001 per 1M tokens

Example

curl https://api.mixen.ai/v1/chat/completions \
  -H "Authorization: Bearer $MIXEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4.1-flash",
    "reasoning_effort": "off",
    "messages": [{"role": "user", "content": "Translate this text into Russian: <text>"}]
  }'

Pricing

Per 1M tokens: input $0.375, output $1.50, cache read $0.008 — the cheapest input among the catalog’s reasoning models. Current prices — catalog.

Strengths and limitations

• A rare combination: controllable reasoning with a real off at a price far below senior tiers.
• Files are not accepted on input; images are.
• For heavy code and analysis the senior tier is DeepSeek V4 Pro.

All models — in the catalog. The general workflow — chat guide.