chooseaimodel
← All models
Flashfastest and cheapest, lighter reasoningOpen Weights

DeepSeek V4 Flash Vision Exp (batch) Pricing

DeepSeek · deepseek-deepseek-v4-flash-vision-exp-batch

ShareXFacebookLinkedIn
Input / 1M tokens
$0.110
Output / 1M tokens
$0.330
Context window
1,048,576 tokens
1,398 pages of text
Max output
943,718 tokens

What it costs in practice

Estimate your own workloadOpens the calculator with DeepSeek V4 Flash Vision Exp (batch) preloaded — adjust volume and token counts there.

Price history

Only one price point recorded so far.

The staircase chart appears once a price change is detected.

What it excels at

Handles large-batch inference over long multimodal contexts at low per-token cost. Open weights enable local or custom deployment for volume workloads.

The business tradeoff

Experimental release shows variable output stability. Self-hosting introduces latency and infrastructure overhead not present in managed APIs.

Cheaper from DeepSeek

Same tier from other providers

Head-to-head

Vendor list rates, as of Sep 8, 2026 · source: openrouter · per-request examples assume 1,200 input + 400 output tokens.