← Model list

DeepSeek-R1

deepseek·deepseek/deepseek-r1·Deprecated·Open weights·deepseek-thinking series

Classic open reasoning model for transparent math, coding, and deliberate problem solving

At a glance
Reference price$3.00 / $7.00 per 1M
Cache read · Abacus
Lowest paid$0.40 / $1.70
NanoGPT Gateway · 7.5× spread
1 more $0 channels
Context128,000
⚠ Providers report 64,000–163,840; the table below is authoritative
Output limit8,192
Capabilities
ReasoningTool useStructured outputTemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
Text
Knowledge cutoff2024-07
Released / updated2025-01-20 / 2025-05-29

Quality & performanceArtificial Analysis · Intelligence Index v4.1

Intelligence18.6
Value26
Coding24.6
Agentic3.1
Output speed
TTFT
Cost per task$0.3241
Value formulaIQ 18.6 ÷ min blended $0.725 = 26

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 18 providers15 with public prices · 1 free

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
iFlowGatewayFree128,00032,000
NanoGPTGateway$0.40$1.70$0.20128,0008,192
D.Run (China)Gateway$0.55$2.20131,07232,000
Alibaba (China)First-party$0.574$2.29131,07216,384
Eden AI
deepinfra/deepseek-ai/DeepSeek-R1
Gateway$0.70$2.40163,84032,768
Hugging Face
deepseek-ai/DeepSeek-R1
Cloud$0.70$2.5064,00032,768
OpenRouterGateway$0.70$2.5064,00016,000
Kilo GatewayGateway$0.70$2.5064,00016,000
Amazon Bedrock
deepseek.r1-v1:0
Cloud$1.35$5.40128,00032,768
Merge GatewayGateway$1.35$5.40163,84040,960

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

1NanoGPT$141.00
2D.Run (China)$220.00
3Alibaba (China)$229.50
4Eden AI$260.00
5Hugging Face$265.00
6OpenRouter$265.00

1 more channels offer $0 (iFlow); free tiers usually have rate limits and no SLA, excluded from ranking.

The cheapest paid channel is the only channel.

Benchmark4 items

NameConditionsScoreMetricSource
Aider Polyglot56.9percent correctSource ↗
Artificial Analysis Coding Index15.9indexSource ↗
SciCode35.7percent correctSource ↗
Terminal-Bench Hard6.1success rateSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Reasoning control

effort = high

17 / 18 providers expose no reasoning control (reasoning_options: []).

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$7.00$02026-08-052026-08-132026-08-05 · Input list $3.002026-08-06 · Input list $3.002026-08-07 · Input list $3.002026-08-08 · Input list $3.002026-08-09 · Input list $3.002026-08-10 · Input list $3.002026-08-11 · Input list $3.002026-08-12 · Input list $3.002026-08-13 · Input list $3.002026-08-05 · Output list $7.002026-08-06 · Output list $7.002026-08-07 · Output list $7.002026-08-08 · Output list $7.002026-08-09 · Output list $7.002026-08-10 · Output list $7.002026-08-11 · Output list $7.002026-08-12 · Output list $7.002026-08-13 · Output list $7.002026-08-05 · Min blended $0.7252026-08-06 · Min blended $0.7252026-08-07 · Min blended $0.7252026-08-08 · Min blended $0.7252026-08-09 · Min blended $0.7252026-08-10 · Min blended $0.7252026-08-11 · Min blended $0.7252026-08-12 · Min blended $0.7252026-08-13 · Min blended $0.725