DeepSeek vs Claude API pricing. Side by side.

DeepSeek’s open-weight V4 models and Anthropic’s Claude, matched tier by tier, with each lab’s list price and what the same tokens cost on Zurelay, where both run on one API key.

DeepSeek models
5
Claude models
10
Below list
43–90%
Get your API keyCompare by tier

By tier

Matched tier by tier. As each lab describes them.

Built for coding and agents

DeepSeek V4 Pro is DeepSeek’s largest model, at 1.6 trillion parameters, built for coding and agents. Anthropic built Opus 5.5 for long-running agentic coding.

DeepSeek V4 Pro 0813
DeepSeek
84%off
Input / 1M
$0.23 $1.32
Output / 1M
$0.62 $3.96
Context
1M tokens
10K requests
$7.70
Reads
Text
Claude Opus 5.5
Anthropic
65%off
Input / 1M
$1.40 $4.00
Output / 1M
$7.00 $20.00
Context
1M tokens
10K requests
$63.00
Reads
Text, images and PDFs

For the typical load: DeepSeek V4 Pro 0813 costs 88% less than Claude Opus 5.5 ($7.70 against $63.00).

Fast and low-cost

V4.1 Flash is DeepSeek’s newest model, the one its own API now serves for Flash requests. Haiku 5.5 is Anthropic’s fastest and lowest-priced model.

DeepSeek V4.1 Flash
DeepSeek
90%off
Input / 1M
$0.03 $0.30
Output / 1M
$0.12 $1.20
Context
1M tokens
10K requests
$1.20
Reads
Text and images
Claude Haiku 5.5
Anthropic
60%off
Input / 1M
$0.04 $0.10
Output / 1M
$0.20 $0.50
Context
1M tokens
10K requests
$1.80
Reads
Text, images and PDFs

For the typical load: DeepSeek V4.1 Flash costs 33% less than Claude Haiku 5.5 ($1.20 against $1.80).

The typical load is 10,000 requests of 2,000 tokens in and 500 out. Struck-through prices are each lab’s list price.

Every model

Every DeepSeek and Claude model. Cheapest first.

DeepSeek and Claude API prices on Zurelay and at each lab's list price, cheapest first
ModelContextInput / 1MOutput / 1M10K requestsYou save
DeepSeek V4 Flash 0731DeepSeek · deepseek-v4-flash-07311M
$0.029
$0.076
$0.087
$0.153
$1.015
$2.285
43%off
DeepSeek V4.1 Flash 1017DeepSeek · deepseek-v4.1-flash-10171M
$0.03
$0.30
$0.12
$1.20
$1.20
$12.00
90%off
DeepSeek V4.1 FlashDeepSeek · deepseek-v4.1-flash1M
$0.03
$0.30
$0.12
$1.20
$1.20
$12.00
90%off
DeepSeek V4 FlashDeepSeek · deepseek-v4-flash1M
$0.03
$0.30
$0.12
$1.20
$1.20
$12.00
90%off
Claude Haiku 5.5Anthropic · claude-haiku-5-51M
$0.04
$0.10
$0.20
$0.50
$1.80
$4.50
60%off
DeepSeek V4 Pro 0813DeepSeek · deepseek-v4-pro-08131M
$0.23
$1.32
$0.62
$3.96
$7.70
$46.20
84%off
Claude Haiku 4.5Anthropic · claude-haiku-4-5200K
$0.54
$1.00
$2.70
$5.00
$24.30
$45.00
46%off
Claude Sonnet 5.5Anthropic · claude-sonnet-5-51M
$0.90
$2.00
$4.50
$10.00
$40.50
$90.00
55%off
Claude Opus 5.5Anthropic · claude-opus-5-51M
$1.40
$4.00
$7.00
$20.00
$63.00
$180.00
65%off
Claude Opus 5Anthropic · claude-opus-51M
$1.75
$5.00
$8.75
$25.00
$78.75
$225.00
65%off
Claude Opus 4.7Anthropic · claude-opus-4-71M
$1.78
$5.00
$8.93
$25.00
$80.25
$225.00
64%off
Claude Opus 4.8Anthropic · claude-opus-4-81M
$2.40
$5.00
$12.00
$25.00
$108.00
$225.00
52%off
Claude Opus 4.6Anthropic · claude-opus-4-61M
$2.40
$5.00
$12.00
$25.00
$108.00
$225.00
52%off
Claude Fable 5.1Anthropic · claude-fable-5-11M
$3.50
$10.00
$17.50
$50.00
$157.50
$450.00
65%off
Claude Fable 5Anthropic · claude-fable-51M
$3.50
$10.00
$17.50
$50.00
$157.50
$450.00
65%off

Prices in US dollars per 1M tokens, live from the catalog, with each lab's list price struck through. 10K requests is 10,000 requests of 2,000 tokens in and 500 out, at each price. You save is on output tokens.

Try both

One key for both. Test on your own work.

Benchmarks don’t tell you which model handles your prompts better. Run the same requests through both and compare the answers, the tokens and the bill.

On Zurelay, every DeepSeek and Claude model is on the same key and the same OpenAI-compatible endpoint, so switching is a change of model name. Claude also answers on Anthropic’s Messages API, for Claude Code and the Anthropic SDKs.

compare.py
from openai import OpenAI
client = OpenAI(base_url="https://api.zurelay.com/v1", api_key="YOUR_ZURELAY_KEY")
prompt = [{"role": "user", "content": "Refactor this function: ..."}]
# Same key, same code: only the model name changes
for model in ["deepseek-v4.1-flash", "claude-haiku-5-5"]:
    reply = client.chat.completions.create(model=model, messages=prompt)
    print(model, reply.usage.total_tokens, reply.choices[0].message.content[:200])

FAQ

DeepSeek vs Claude questions.

Anything else? Ask our team and we’ll answer by email.

Is DeepSeek or Claude cheaper?

It depends on the tier. For 10,000 requests of 2,000 tokens in and 500 out on Zurelay: built for coding and agents, DeepSeek V4 Pro 0813 costs 88% less than Claude Opus 5.5 ($7.70 against $63.00); fast and low-cost, DeepSeek V4.1 Flash costs 33% less than Claude Haiku 5.5 ($1.20 against $1.80). Both are 43–90% below their labs' list prices.

Can I use DeepSeek and Claude with one API key?

Yes. One Zurelay key calls every DeepSeek and Claude model. With an OpenAI SDK, point it at https://api.zurelay.com/v1 and change only the model name to switch between them. Claude also answers on Anthropic's Messages API, for Claude Code and the Anthropic SDKs.

Which is better for coding?

DeepSeek's V4 models are open-weight models built for coding and agents, and Anthropic describes Claude Opus 5.5 as built for long-running agentic coding. DeepSeek costs far less per token; whether it does your work as well is a question for your own tests. On Zurelay, running both is a change of model name.

Are these the real DeepSeek and Claude models?

Yes. Each model is served over several routes run by independent sellers, and every route is tested before it serves requests and every day after, with identity, dated-knowledge and reasoning checks that catch a different or older model sold under the name you asked for.

More pricing

Stop paying list price.
Start saving today.

Create a key in seconds, point the SDK you already use at Zurelay, and every request costs up to 90% less from the first token.

Get your API keyTalk to an engineer