Appndo
All models

DeepSeek V3.1

DeepSeekOpen weights164K contextReasoningTool calling

LMArena score 1420 · #116 overall · #150 in coding (2026-10-13)

Lowest price, from DeepInfra

Pricing page

Per million tokens

Input
$0.25
Output
$0.95
Cached input
$0.216
Batch input / output
—
  • Up to 164K output tokens per request.

Estimate your monthly cost

Est. per month = input tokens × input price + output tokens × output price. Prices are per million tokens, so 10M input tokens at $3.00 cost $30.

From DeepInfra: $4.40 a month.

Where to buy it (7)

Sorted by your monthly estimate. Prices per million tokens.

  • DeepInfraCheapest

    Input / output
    $0.25 / $0.95
    Context
    164K
    Per month
    $4.40

    deepinfra/deepseek-ai/DeepSeek-V3.1

  • Novita AI

    Input / output
    $0.27 / $1.00
    Context
    131K
    Per month
    $4.70

    novita/deepseek/deepseek-v3.1

  • Baseten

    Input / output
    $0.50 / $1.50
    Context
    —
    Per month
    $8.00

    baseten/deepseek-ai/DeepSeek-V3.1

  • W&B Inference

    Input / output
    $0.55 / $1.65
    Context
    161K
    Per month
    $8.80

    wandb/deepseek-ai/DeepSeek-V3.1

  • Google Vertex AI

    Input / output
    $0.60 / $1.70
    Context
    164K
    Per month
    $9.40

    vertex_ai/deepseek-ai/deepseek-v3.1-maas

  • Replicate

    Input / output
    $0.672 / $2.02
    Context
    164K
    Per month
    $10.75

    replicate/deepseek-ai/deepseek-v3.1

  • SambaNova

    Input / output
    $3.00 / $4.50
    Context
    131K
    Per month
    $39.00

    sambanova/DeepSeek-V3.1

Cloud and hosted sellers may add regional pricing, minimums or platform fees. Supports: Reasoning, Tool calling.

Similar models