Skip to content
CostPerPrompt

R1 0528 API Pricing

DeepSeek · context window 164K · prices updated 2026-09-15

Input per 1M tokens
$0.5
Output per 1M tokens
$2.15
Cached input per 1M
$0.35
30% cheaper than fresh input

Where R1 0528 sits on price

At $2.15 per million output tokens, R1 0528 is a mid tier model — right at the median output price of $2.08, which puts it cheaper than 50% and pricier than 50% of the models we track. It sits 15th cheapest of the 18 DeepSeek models we track. Output costs 4.3× input, the usual spread — trim both, starting with the answer length.

Output tokens per $1
465,116
One full context fill
$0.0819
Cheaper than
50% of tracked models

Cached input is 30% cheaper. On an input-heavy workload you need roughly a 83% 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 R1 0528

These three workloads are the ones teams actually run on a mid-priced model — a $2.15/1M model is not bought for the same job as one ten times the price.

Workload Per request Per month
Support chatbot — 500 conversations/day (5K in / 1.4K out) Conversation history re-sent each turn — the classic prompt-caching win. $0.0055 $82.65
Document pipeline — 10K docs/day (3K in / 500 out) Batch-friendly: no user waiting, so a ~50% batch discount usually applies. $0.0026 $773
Small app — 1K requests/day (500 in / 150 out) The baseline most side projects actually run at. $0.0006 $17.18

Model your exact traffic in the API cost calculator — it preloads R1 0528 with caching and batch options.

R1 0528 price history

R1 0528's price has not moved since we began tracking it on Aug 2, 2026 — 44 days of stability in a market where 78 of the 334 models we track have repriced over the same period, including 12 of DeepSeek's own 18 models.

Change-points from our daily price snapshots (tracking since Aug 2, 2026; intraday moves between snapshots are not captured).

R1 0528 vs Qwen3 VL 8B Thinking

The closest-priced alternative from another vendor is Qwen3 VL 8B Thinking (Alibaba (Qwen)) — 2% cheaper on output, with $0.32 less 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 Qwen3 VL 8B Thinking pricing →

Cheaper alternatives

More DeepSeek models

Frequently asked questions

How much does the R1 0528 API cost?

R1 0528 costs $0.5 per million input tokens and $2.15 per million output tokens, with cached input at $0.35 per million (30% cheaper). That works out to roughly 465,116 output tokens per dollar.

What does the support chatbot workload cost on R1 0528?

Support chatbot — 500 conversations/day (5K in / 1.4K out) costs about $0.0055 per request and $82.65 per month on R1 0528. Conversation history re-sent each turn — the classic prompt-caching win.

What does it cost to fill R1 0528's 164K context window?

Sending 164K of input in a single request costs $0.0819 at $0.5 per million tokens — before any output. With prompt caching that same fill drops to about $0.0573 on repeat requests. This is why large context windows are cheap to advertise and expensive to actually use.

Is R1 0528 worth the price?

R1 0528 sits in the middle of the market (15 of 18 by price within DeepSeek). The honest test is a routing experiment: send the same 200 real requests to this model and to a tier below, and compare failure rate against the price gap — most teams find a majority of traffic never needed the pricier model.