DeepSeek: R1

deepseek · Released Jan 20, 2025
DFO score #297 / 749
12.4 leaderboard score
Speed #62 / 355
174.6 tok/s
Input Price #529 / 749
$0.700 per 1M tokens
Output Price #527 / 749
$2.50 per 1M tokens
Context #556 / 749
64,000 tokens

Analysis Summary

DeepSeek R1 is a text-only reasoning model released in January 2025, with a 64K context window and support for tools and function calling. Its measured reasoning capability exceeds its coding and agentic results, which are limited for advanced software engineering or autonomous multi-step execution. No vision support is listed, so its input remains text-focused.

Its pricing, $0.70 per million input tokens and $2.50 per million output tokens, supports selective use for reasoning-heavy tasks, but output costs make it less attractive for high-volume generation. It may fit structured analysis and tool-assisted workflows that have clear steps and human oversight. It is not a strong choice for long codebases or agents that need dependable terminal execution based on the supplied data.

Pilot it on bounded reasoning and extraction tasks, with tool-call checks and review of consequential answers. Use a stronger coding or agentic model when software changes or autonomous completion are central to the job.

Assessed September 23, 2026

Editorial notes

DeepSeek R1 combines a 64K context window with tool and function calling at low listed prices. Its coding and agentic results remain limited, so it fits bounded reasoning tasks better than autonomous engineering workflows.

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 R1 combines a 64K context window with tool and function calling at low listed prices. Its coding and agentic results remain limited, so it fits bounded reasoning tasks better than autonomous engineering workflows.

#297 of 749 overall Down 90 this week

Benchmarks

Intelligence Index 8.4 #199 of 454 · Artificial Analysis
Coding Index 15.7 #126 of 199 · Artificial Analysis
Agentic Index 0.7 #157 of 181 · Artificial Analysis
Creative Writing Not rated Arena rating
Instruction Following Not rated Arena rating

Artificial Analysis data refreshed Nov 16, 2026. Arena ratings from the Nov 9, 2026 text leaderboard, with style control.

Artificial Analysis no longer lists this model, so its indexes are converted from their older scale to stay comparable.

How DeepSeek: R1 compares

DeepSeek: R1 ranks #199 of 454 AI models we track for overall intelligence, #126 of 199 for coding, #157 of 181 for agentic tasks. Its 64K-token context window is larger than 26% of the models we list. At $0.70 per million input tokens it is cheaper than 29% of comparable models.

Position in the field
DFO score: ahead of 60% of models #297
Speed: faster than 83% of models #62
Price: cheaper than 29% of models #529
Context: larger than 26% of models #556
worst in fieldmedianbest in field
Price vs frontier peers · $ per 1M tokens
DeepSeek: R1 $0.70 in $2.50 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 $0.70 $0.000700
Output $2.50 $0.002500

What would DeepSeek: R1 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

$4.46/mo DeepSeek: R1 $0.0015 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 749 models → Price Calculator

DFO AI AUTOMATION

These numbers get smaller with the right architecture.

We route routine calls to cheap models and save DeepSeek: R1 for the hard ones. Most clients cut their estimate by 60-80%.

Talk to our team

About DeepSeek: R1

DeepSeek R1 is here: Performance on par with OpenAI o1, but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass..

Frequently asked questions about DeepSeek: R1

How much does DeepSeek: R1 cost?

DeepSeek: R1 costs $0.70 per million input tokens and $2.50 per million output tokens.

What is the context window of DeepSeek: R1?

DeepSeek: R1 has a context window of 64,000 tokens (64K).

Is DeepSeek: R1 good for coding?

On our coding benchmark index, DeepSeek: R1 ranks #126 of 199 models, placing it in the broader range of the field for code generation and debugging.

What can DeepSeek: R1 do?

DeepSeek: R1 supports tool use and function calling.

Who created DeepSeek: R1?

DeepSeek: R1 is developed by DeepSeek and was released on January 20, 2025.

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