DeepSeek · Flagship · 1M context window
$0.0129 per call at 15,000 input / 1,500 output tokens.
Per call
$0.0129
10K calls/mo
$128.7
Cache savings
97% off input
Batch savings
0% off all
Adjust token counts, volume, and discount toggles to model your pdf summarization spend. Prices verified daily against provider pricing pages.
Call DeepSeek V4 Pro with the OpenAI Python SDK pointed at api.deepseek.com:
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.deepseek.com/v1",
api_key=os.environ["DEEPSEEK_API_KEY"],
)
response = client.chat.completions.create(
model="deepseek-v4-pro",
max_tokens=1500,
messages=[{"role": "user", "content": "Your prompt here"}]
)
# Cost: ~$0.0129 per call at standard rates
# Input: $0.66/MTok
# Output: $1.98/MTokPer-call cost for all models scoring 50+ on pdf summarization benchmarks. Hover for details.
Cheapest viable option: Gemini 3 Flash at $0.0120/call (7% less, 83/100 quality).
PDFs are the worst case for input cost: extracted text carries page furniture, headers, and footnotes that inflate token counts well past what the content warrants. Cleaning the extraction before sending it often cuts more cost than changing models. Long PDFs also routinely cross long-context pricing thresholds, so chunking is both a quality and a cost decision.
For DeepSeek V4 Pro specifically, that means output tokens cost 3.0x more than input, at $1.98 vs $0.66 per MTok. At this use case’s 15,000/1,500 token split, output accounts for 23% of the $0.0129 per-call cost.
| Direction | $/MTok | Tokens per call |
|---|---|---|
| Input | $0.66 | 15,000 |
| Output | $1.98 | 1,500 |
Caching drops input cost from $0.66 to $0.02/MTok. For repeated system prompts, that saves $0.009573 per call.
Submit requests asynchronously and receive 0% off all token costs. This works best for offline pipelines where latency requirements are flexible.
| Monthly volume | Cost per call | Standard monthly | Cached monthly | You save with caching |
|---|---|---|---|---|
| 10,000 calls | $0.0129 | $128.7 | $32.97 | $95.73 |
| 50,000 calls | $0.0129 | $643.5 | $164.84 | $478.67 |
| 100,000 calls | $0.0129 | $1,287 | $329.67 | $957.33 |
| 500,000 calls | $0.0129 | $6,435 | $1,648.35 | $4,786.65 |
| 1,000,000 calls | $0.0129 | $12,870 | $3,296.7 | $9,573.3 |
Benchmark data for pdf summarization is pending for DeepSeek V4 Pro. Test against your own evaluation suite before moving to production.
Benchmark data is not yet available for DeepSeek V4 Pro. We recommend testing against your own evaluation criteria.
10 models score 80+ on pdf summarization benchmarks, sorted by input rate. Cheapest: Gemini 3 Flash (7% less, 83/100 quality).
| Model | Provider | Input $/MTok | Output $/MTok | Quality | Per call | vs DeepSeek |
|---|---|---|---|---|---|---|
| Gemini 3 Flash | Gemini | $0.50 | $3.00 | 83/100 | $0.0120 | 7% cheaper |
| Gemini 3.6 Flash | Gemini | $0.75 | $3.75 | 75/100 | $0.0169 | 31% more |
| Gemini 3.7 Flash | Gemini | $0.75 | $3.75 | 79/100 | $0.0169 | 31% more |
| Gemini 3.8 Flash | Gemini | $0.75 | $3.75 | 81/100 | $0.0169 | 31% more |
| Claude Haiku 4.5 | Claude | $1.00 | $5.00 | 82/100 | $0.0225 | 75% more |
| Gemini 3.5 Flash | Gemini | $1.50 | $9.00 | 76/100 | $0.0360 | 180% more |
| Gemini 3.1 Pro | Gemini | $2.00 | $12.00 | 80/100 | $0.0480 | 273% more |
| Grok 4.7 | Grok | $2.00 | $6.00 | 75/100 | $0.0390 | 203% more |
| Claude Fable 5 | Claude | $10.00 | $50.00 | 77/100 | $0.2250 | 1648% more |
| GPT 6 Astra | GPT | $10.00 | $50.00 | 76/100 | $0.2250 | 1648% more |
Provider
DeepSeek
Tier
Flagship
Context
1M tokens
Released
Not published
Prices are off-peak rates from api-docs.deepseek.com (read 2026-10-11). Peak-hour prices are 2x the listed input and output rates — Tokeven does not model time-of-day tiers; off-peak is used as the base. Cache hit: $0.022/MTok input. No batch discount published. Not yet benchmarked, so never recommended.
| Model | Tier | Input $/MTok | Output $/MTok | Context |
|---|---|---|---|---|
| DeepSeek V3 | Mid-tier | $0.27 | $1.10 | 164K |
| DeepSeek R1 | Mid-tier | $0.55 | $2.19 | 128K |
| DeepSeek Flash | Fast / Budget | $0.15 | $0.60 | 1M |
You can compare DeepSeek V4 Pro head-to-head against any other model on our model comparison pages, or track how pricing has changed over time on the pricing history tracker.
See how DeepSeek V4 Pro costs change across different workload types.
Code Generation
$0.009240/call
Unit Test Generation
$0.0119/call
SQL Generation
$0.002970/call
API Generation
$0.0139/call
Code Refactoring
$0.0106/call
Code Review
$0.007260/call
PR Review
$0.0112/call
Security Review
$0.008910/call
Bug Detection
$0.007260/call
Document Summarization
$0.008580/call
Meeting Notes
$0.009240/call
Email Summarization
$0.002970/call
Chatbot
$0.003300/call
RAG Pipeline
$0.006930/call
Customer Support Bot
$0.003960/call
FAQ Bot
$0.002574/call
Data Extraction
$0.004620/call
JSON Extraction
$0.003564/call
Invoice Parsing
$0.002508/call
Resume Parsing
$0.003234/call
Chain of Thought
$0.0119/call
Legal Analysis
$0.0145/call
Math Reasoning
$0.006930/call
Financial Analysis
$0.0132/call
Content Moderation
$0.000528/call
Sentiment Analysis
$0.000528/call
Categorization
$0.000825/call
Intent Detection
$0.000396/call
Copywriting
$0.004290/call
Blog Writing
$0.0165/call
Marketing Copy
$0.006468/call
Product Descriptions
$0.003300/call
A typical pdf summarization call using DeepSeek V4 Pro costs about $0.0129 per call. This estimate uses 15,000 input tokens and 1,500 output tokens at standard rates. Standard rates are $0.66 per million input tokens and $1.98 per million output tokens.
Our benchmark data does not yet include a pdf summarization ranking for DeepSeek V4 Pro. We recommend testing it against your own evaluation criteria before production usage.
The most affordable model for pdf summarization above 80/100 quality is Gemini 3 Flash. It comes from Gemini at $0.50 per million input tokens. Output costs $3.00 per million tokens with a quality score of 83/100. Switching from DeepSeek V4 Pro to Gemini 3 Flash saves about 7% per API call.
Three strategies lower DeepSeek V4 Pro costs for this task type effectively. Prompt caching gives a 97% discount on input tokens when reusing system prompts. The batch API gives 0% off all tokens for asynchronous processing workloads. Prompt optimization reduces token counts through more concise instructions and context windows. Combining caching with batching brings your cost from $0.0129 to about $0.000425 per call.
DeepSeek has 3 other models alongside DeepSeek V4 Pro today: DeepSeek V3 at $0.27/$1.10 per million tokens (mid-tier), DeepSeek R1 at $0.55/$2.19 per million tokens (mid-tier), DeepSeek Flash at $0.15/$0.60 per million tokens (fast / budget). The right choice depends on whether your pdf summarization workload prioritizes quality, cost, or speed.
DeepSeek V4 Pro supports a context window of 1M tokens per API call. This sets the maximum combined length of your input prompt and generated response. For pdf summarization, typical calls use 15,000 input and 1,500 output tokens. That is well within this limit, leaving room for longer prompts or multi-turn conversations.
This calculator estimates costs based on published rates only. Connect your API usage to see where your money goes and which models drive your spend.