DeepSeek: DeepSeek V4 Pro

deepseek · Released Apr 24, 2026
DFO score #60 / 753
34.0 leaderboard score
Speed #197 / 358
89.8 tok/s
Input Price #560 / 753
$0.900 per 1M tokens
Output Price #468 / 753
$1.50 per 1M tokens
Context #39 / 753
1M tokens

Analysis Summary

DeepSeek V4 Pro is a DeepSeek model released in April 2026, with a 1,048,576-token context window, tool use, and function calling. Its coding index is 68.8, while its agentic index is 41.3. The supplied results also show strong instruction following, long-context performance, and task success on the tau benchmark, making it a capable choice for multi-step work over large inputs.

For agencies, the combination suits software engineering assistance, codebase and document analysis, and tool-driven workflows where following instructions and retaining context matter. The provided modality is text only, so image-based work is not established. Pricing is $0.9464 per million input tokens and $1.8928 per million output tokens, which is a meaningful trade-off for high-volume use. Adopt it for complex jobs where coding, context, or tool execution can justify the spend, and route routine traffic to a cheaper model.

Assessed September 23, 2026

Editorial notes

DeepSeek V4 Pro pairs strong coding and tool-use results with a 1M-token context, function calling, and strong instruction following; its premium pricing favors complex, high-value workloads.

The DFO score ranks models on the Artificial Analysis Intelligence Index and is refreshed as new models launch. The write-up and verdict are our own. Feedback?

DFO Verdict

DeepSeek V4 Pro pairs strong coding and tool-use results with a 1M-token context, function calling, and strong instruction following; its premium pricing favors complex, high-value workloads.

#60 of 753 overall Down 5 this week

Benchmarks

Intelligence Index 23.1 #60 of 456 · Artificial Analysis
Coding Index 68.8 #13 of 199 · Artificial Analysis
Agentic Index 41.3 #14 of 181 · Artificial Analysis
Creative Writing 1,447 #12 of 32 · Arena rating
Instruction Following 1,453 #18 of 35 · Arena rating
GPQA Diamond 88.8%
HLE 35.9%
SciCode 50%
TerminalBench Hard 46.2%
τ²-Bench 96.2%
IFBench 76.5%
LCR 66.3%

Artificial Analysis data refreshed Nov 17, 2026. Arena ratings from the Nov 9, 2026 text leaderboard, with style control. Bars: magenta = reasoning, ink = coding and agents, cyan = instructions and long context.

How DeepSeek: DeepSeek V4 Pro compares

DeepSeek: DeepSeek V4 Pro ranks #60 of 456 AI models we track for overall intelligence, #13 of 199 for coding, #14 of 181 for agentic tasks. Its 1M-token context window is larger than 95% of the models we list. At $0.90 per million input tokens it is cheaper than 26% of comparable models.

Position in the field
DFO score: ahead of 92% of models #60
Speed: faster than 45% of models #197
Price: cheaper than 26% of models #560
Context: larger than 95% of models #39
worst in fieldmedianbest in field
Price vs frontier peers · $ per 1M tokens
DeepSeek: DeepSeek V4 Pro $0.90 in $1.50 out
Anthropic: Claude Fable 5.2 $8.00 in $40.00 out
Google: Gemini 4 Carbon $2.00 in $10.00 out
Google: Gemini 4 Argon $2.00 in $10.00 out

Dark bar = input · light bar = output, scaled to the priciest peer.

Context window vs peers · tokens

1M tokens ≈ 8 full-length novels or ~2,500 pages of business documents in a single request.

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.90 $0.000900
Output $1.50 $0.001500

What would DeepSeek: DeepSeek V4 Pro cost your business?

Pick the job that looks most like yours, then fine-tune with the sliders. Estimates update live.

A website chatbot handling around 100 customer conversations a day, a few short messages each.

3,000
One request is one message, email, draft or automation call.
1,200 tokens

$3.89/mo DeepSeek: DeepSeek V4 Pro $0.0013 per request
$63.36/mo Anthropic: Claude Fable 5.2 $0.02 per request
$15.84/mo Google: Gemini 4 Carbon · best value $0.0053 per request

Full calculator with 753 models → Price Calculator

DFO AI AUTOMATION

These numbers get smaller with the right architecture.

We route routine calls to cheap models and save DeepSeek: DeepSeek V4 Pro for the hard ones. Most clients cut their estimate by 60-80%.

Talk to our team

About DeepSeek: DeepSeek V4 Pro

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,..

Frequently asked questions about DeepSeek: DeepSeek V4 Pro

How much does DeepSeek: DeepSeek V4 Pro cost?

DeepSeek: DeepSeek V4 Pro costs $0.90 per million input tokens and $1.50 per million output tokens.

What is the context window of DeepSeek: DeepSeek V4 Pro?

DeepSeek: DeepSeek V4 Pro has a context window of 1,048,576 tokens (1M).

Is DeepSeek: DeepSeek V4 Pro good for coding?

On our coding benchmark index, DeepSeek: DeepSeek V4 Pro ranks #13 of 199 models, placing it in the top quartile of the field for code generation and debugging.

What can DeepSeek: DeepSeek V4 Pro do?

DeepSeek: DeepSeek V4 Pro supports tool use and function calling.

Who created DeepSeek: DeepSeek V4 Pro?

DeepSeek: DeepSeek V4 Pro is developed by DeepSeek and was released on April 24, 2026.

© 2026 Design for Online Ltd. Registered in England and Wales No. 10328553. VAT Registered. Design for Online® and Forerunner® are registered trademarks.