All models

DeepSeek

DeepSeek R1 balancedruns locally

deepseek/deepseek-r1

Input / 1M

$0.55

USD per million tokens

Output / 1M

$2.19

USD per million tokens

Est. cost / token

$0.00000096

$0.96 per 1M, 3:1 mix

Context window

66K

max output 8K

Major features

CachingReasoningTools

Task suitability (heuristic 1–5, not a benchmark)

  • coding4
  • reasoning4
  • math4
  • summarization4
  • extraction4
  • classification4
  • chat4
  • creative4
  • translation4
  • vision1
  • long context2
  • agentic4

Run it locally

Will it run on my machine? →
ollama run deepseek-r1:671b
QuantizationQualityMemory needed (8K ctx)
Q2_Kpoor303 GB
Q3_K_Mfair352 GB
Q4_K_Mrecommendedgood438 GB
Q5_K_Mvery good509 GB
Q6_Kexcellent585 GB
Q8_0near-lossless756 GB
F16lossless1417 GB

Estimates. Mixture-of-experts: memory covers all parameters; only the active ones are used per token, so it runs faster than its size suggests.

Price history

RecordedInputOutputCachedSource
2026-11-05 03:20$0.55$2.19$0.14litellm

A row is added whenever a sync detects a price change.

Pricing (USD per 1M tokens)

Input
$0.55
Output
$2.19
Cached input
$0.14
Blended (3:1)
$0.96

Details

Company
DeepSeek
Version
1
Family
DeepSeek
Released
—
Parameters
671B (37B active)
Max output
8K
Open weights
Yes
Inputs
text
Last synced
29 min ago