← All models
Flash — fastest and cheapest, lighter reasoningOpen Weights
DeepSeek V4 Flash Vision Exp (batch) Pricing
DeepSeek · deepseek-deepseek-v4-flash-vision-exp-batch
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
Typical request (1,200 in + 400 out tokens)$0.0003
1,000 requests / month$0.264/mo10,000 requests / month$2.64/mo100,000 requests / month$26.40/moEstimate 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
Head-to-head
Vendor list rates, as of Sep 8, 2026 · source: openrouter · per-request examples assume 1,200 input + 400 output tokens.