Skip to main content

DeepSeek R1

DeepSeek R1 is here: Performance on par with OpenAI o1, but open-sourced and with fully open reasoning tokens. $0.70 per million input tokens, $2.50 per million output tokens. 163,840 token context window, maximum output of 16,000 tokens. Higher uptime with 2 providers. Includes independent benchmarks from Artificial Analysis.

Input / output price
$0.70 / $2.50 per 1M
Context
128K tokens
Released
Jan 20, 2025
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "deepseek-r1".

Read docs

Providers

4 providers, cheapest first, priced per 1M tokens after gateway discounts.

ProviderInputOutput
Novitanovita/deepseek-r1
$0.70/M$2.50/M
Bedrockbedrock/deepseek-r1
$1.35/M$5.40/M
Concentrateconcentrate/deepseek-r1
$1.35/M$5.40/M
Azure DeepSeekazure_deepseek/deepseek-r1
$8.00/M$8.00/M

Is a provider down? AWS Bedrock status

Pricing

Averaged across 4 providers, in USD per 1M tokens after gateway discounts.

Average input
$2.85 per 1M
Average output
$5.33 per 1M
Average cache read
$2.00 per 1M

Cost by volume

Total cost from 0 to 1T tokens on the cheapest and most expensive provider.

Highest$8.00/MAzure DeepSeekLowest$0.70/MNovita

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

13tokens/sacross providers

Faster than 40% of the other 596 models on the gateway

Response latency

Lower is better

7.8sacross providers

Faster than 22% of the other 776 models on the gateway

Benchmarks

The Respan Index plus 12 published scores from 7 sources, each linked to where it was published.

18.8/ 100

Respan Index#138 of 176

176 ranked models. The shaded band is this model's 95% interval, 13.4 to 29.5. DeepSeek V3.1 and GPT-5 nano fall inside it, so the gap is within the margin of error.
DeepSeek R1 by category: Coding 3.6, Agents 16.4, Reasoning 21.2, Knowledge & language 17.6, Human preference 45.1
  1. 138DeepSeek R118.8
  2. 137DeepSeek V3.118.9
  3. 138GPT-5 nano18.8

Published scores

BenchmarkScore
SWE-bench Verifiedgithub.com49.2%
WeirdMLHåvard Tveit Ihle36.5%

Compare DeepSeek R1: DeepSeek R1 vs DeepSeek-V4.1-Flash-1017, DeepSeek R1 vs DeepSeek V4.1 Flash, DeepSeek R1 vs DeepSeek V4 Pro, DeepSeek R1 vs DeepSeek V4 Flash, DeepSeek R1 vs DeepSeek-V4-Flash (Preview), DeepSeek R1 vs DeepSeek V3.2

Apps

Use DeepSeek R1 in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

DeepSeek R1 FAQ

Short answers, built from the data on this page.

DeepSeek R1 is a model from Deepseek, released on Jan 20, 2025. DeepSeek R1 is here: Performance on par with OpenAI o1, but open-sourced and with fully open reasoning tokens. $0.70 per million input tokens, $2.50 per million output tokens. 163,840 token context window, maximum output of 16,000 tokens. Higher uptime with 2 providers. Includes independent benchmarks from Artificial Analysis.

On the Respan gateway, DeepSeek R1 costs from $0.70 per 1M input tokens and $2.50 per 1M output tokens on Novita, after gateway discounts. Averaged across its 4 providers, it costs $2.85 per 1M input tokens and $5.33 per 1M output tokens.

DeepSeek R1 is available through 4 providers: Novita, Bedrock, Concentrate, and Azure DeepSeek.

DeepSeek R1 accepts up to 128K tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "deepseek-r1". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

DeepSeek R1 scores 18.8 out of 100 on the Respan Index v1.1, ranking #138 of 176 models. By category: Coding 3.6, Agents 16.4, Reasoning 21.2, Knowledge & language 17.6, and Human preference 45.1. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

DeepSeek R1 has 12 published benchmark scores, including SWE-bench Verified 49.2%, BALROG 34.9%, and GPQA Diamond 71.7%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.