Skip to content
CostPerPrompt

DeepSeek V4.1 Flash 1017 (batch) API Pricing

DeepSeek · context window 1M · prices updated 2026-11-22

Input per 1M tokens
$0.035
Output per 1M tokens
$0.149
Cached input per 1M
$0.0029
92% cheaper than fresh input

Where DeepSeek V4.1 Flash 1017 (batch) sits on price

At $0.149 per million output tokens, DeepSeek V4.1 Flash 1017 (batch) is a budget tier model — 16.8× cheaper than the median output price of $2.50, which puts it in the cheapest 10% of everything we track. It sits 2th cheapest of the 17 DeepSeek models we track. Output costs 4.3× input, the usual spread — trim both, starting with the answer length.

Output tokens per $1
6,711,409
One full context fill
$0.0367
Cheaper than
96% of tracked models

Cached input is 92% cheaper. On an input-heavy workload you need roughly a 27% cache-hit rate to take 25% off the input line — reachable for chatbots and agents that resend the same system prompt and history.

What real workloads cost on DeepSeek V4.1 Flash 1017 (batch)

These three workloads are the ones teams actually run on a budget-priced model — a $0.149/1M model is not bought for the same job as one ten times the price.

Workload Per request Per month
High-volume classification — 200K requests/day (400 in / 20 out) Short output, huge volume: the input price dominates the bill. $0 $102
Small app — 1K requests/day (500 in / 150 out) The baseline most side projects actually run at. $0 $1.20
Long-context analysis — 200 runs/day (250K in / 4K out) Only possible on large-context models; input cost is nearly the whole bill. $0.0093 $56.08

Model your exact traffic in the API cost calculator — it preloads DeepSeek V4.1 Flash 1017 (batch) with caching and batch options.

DeepSeek V4.1 Flash 1017 (batch) price history

DeepSeek last cut DeepSeek V4.1 Flash 1017 (batch)'s output price on Nov 20, 2026 (-1% per output token), from $0.036/$0.15 to $0.035/$0.149 per 1M in/out. That is one of 5 repricings in the 14 days we have tracked it — worth knowing before you hard-code today's rate into a budget.

Date Input /1M Output /1M Output change
Nov 8, 2026 (tracking began) $0.011 $0.2 —
Nov 12, 2026 $0.05 $0.15 -25%
Nov 15, 2026 $0.036 $0.318 +112%
Nov 16, 2026 $0.035 $0.311 -2%
Nov 17, 2026 $0.036 $0.15 -52%
Nov 20, 2026 $0.035 $0.149 -1%

Change-points from our daily price snapshots (tracking since Nov 8, 2026; intraday moves between snapshots are not captured).

DeepSeek V4.1 Flash 1017 (batch) vs Command R7B (12-2024)

The closest-priced alternative from another vendor is Command R7B (12-2024) (Cohere) — priced within a rounding error on output, with $0.0025 more per million input tokens. When two models land this close on price, the decision is quality on your own workload, not the price sheet: run 50 real requests through both and compare.

See Command R7B (12-2024) pricing →

Cheaper alternatives

More DeepSeek models

Frequently asked questions

› How much does the DeepSeek V4.1 Flash 1017 (batch) API cost?

DeepSeek V4.1 Flash 1017 (batch) costs $0.035 per million input tokens and $0.149 per million output tokens, with cached input at $0.0029 per million (92% cheaper). That works out to roughly 6,711,409 output tokens per dollar.

› What does the high-volume classification workload cost on DeepSeek V4.1 Flash 1017 (batch)?

High-volume classification — 200K requests/day (400 in / 20 out) costs about $0 per request and $102 per month on DeepSeek V4.1 Flash 1017 (batch). Short output, huge volume: the input price dominates the bill.

› What does it cost to fill DeepSeek V4.1 Flash 1017 (batch)'s 1M context window?

Sending 1M of input in a single request costs $0.0367 at $0.035 per million tokens — before any output. With prompt caching that same fill drops to about $0.003 on repeat requests. This is why large context windows are cheap to advertise and expensive to actually use.

› Is DeepSeek V4.1 Flash 1017 (batch) worth the price?

DeepSeek V4.1 Flash 1017 (batch) is in the cheapest end of the market — cheaper than 96% of the models we track. At this price the question is not cost but capability: it is a good fit for classification, extraction, routing and bulk summarisation, and a poor fit for multi-step reasoning where a wrong answer costs more than the tokens saved.