Provisional fares. Beta pricing. Your fare is fixed when you submit a batch. View pricing details

DeepSeek V4.1 Flash

Text generation: classify, extract, translate, summarise. Sold per 1,000 weighted tokens: output tokens plus input tokens at 0.25, the published price ratio, 100 token minimum per file. Runs on Google Cloud Vertex AI, not our GPUs: the prompt file leaves our storage.

List fare, per 1,000 tokens: $0.0015
Launch fare, from 20,000 tokens: $0.0009
Minimum, per submission: $0.0075
1,000,000 tokens, one batch: $1.50; $0.90 at the launch fare

The minimum stops mattering at 5,000 tokens. Below that, a submission pays $0.0075 whatever its size. A file under 100 tokens bills as 100 tokens.

Cold start: expect 10 to 30 s, occasionally minutes. This is a queue, and later items in a batch ride on the same start.

Try it in the playground Get an API key

Using it.

Send the prompt as a UTF-8 text/plain file, or as JSON: a messages array of system, user and assistant texts, at most 64.
Units are 1,000 weighted tokens: output tokens, plus input tokens at a quarter, from the answer's own usage. Set max_units to cover max_tokens plus a quarter of the prompt; an answer over it fails as too_many_units, at no charge.
With output json the answer comes as {parsed, data} or, if it could not be parsed, {parsed: false, text}, with truncated set when it stopped at max_tokens.
This model runs on Google Cloud Vertex AI, not our GPUs: the prompt is sent to Google for each run.

Table 1 Limits
Model ID	`deepseek-v41-flash`
Input types	`text/plainapplication/json`
Largest file	0 MB
Largest item	32,000 tokens
Default `max_units`	8
Smallest `max_units`	0.1
Licence	See Vertex AI Model Garden, publisher deepseek-ai

Table 2 Parameters, all optional
Name	Type	Allowed	Default
`max_tokens`	integer	1 to 8,192	2048
`output`	string	text, json	text

Dry run, which writes nothing

curl -s https://www.y-pad.com/v1/batches \
  -H "Authorization: Bearer $YPAD_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v41-flash",
    "items": [{"file": "FILE_ID", "params": {"max_tokens":1024,"output":"text"}}],
    "dry_run": true
  }'

The first-run guide covers the upload before this and the rest of the batch after it.

Who charges less.

Table 3 DeepSeek V4.1 Flash, our launch fare first, then lowest first, per 1,000 audio minutes
Who	Published price	Per 1,000	Against our fares	Checked	Source
Y-PAD, launch fare	$0.0009 per 1,000 tokens$0.0075 minimum per submission, 100 token minimum per file	$0.90	Ours	2 Nov 2026	Fares
DeepSeek	$0.30 per 1M input, $1.20 per 1M output, peakpublished price, weighted the same way	$1.20	Less than our list fare, more than launch	2 Nov 2026	api-docs.deepseek.com
Y-PAD, list fare	$0.0015 per 1,000 tokens$0.0075 minimum per submission, 100 token minimum per file	$1.50	Ours	2 Nov 2026	Fares