DeepSeek V4 Flash Vision Exp

DeepSeek-V4-Flash-Vision-Exp is an experimental multimodal model that combines the agentic reasoning, coding, and world knowledge capabilities of DeepSeek-V4-Flash with advanced visual understanding. It delivers major gains on vision-dependent agent workflows while performing strongly across leading coding, repository, data science, automation, and multimodal benchmarks.

Input / 1M tokens$0.216

Output / 1M tokens$0.647

Maximum context1M

Inference providers3

Price comparison

USD per one million tokens, using current provider data.

Provider latency

Last-hour p50 in milliseconds; lower is better.

Provider uptime

Reported availability over the last 24 hours; not a contractual SLA.

Inference providers

Compare current pricing, performance, and privacy across every available provider.

Deepinfra

99.96%

Context1M

Max output1M

p50 latency466 ms

Throughput144.0 t/s

Input price$0.216

Output price$0.647

Zero retentionNo trainingCaching

Fireworks

100.00%

Context1M

Max output1M

p50 latency1,491 ms

Throughput191.5 t/s

Input price$0.22

Output price$0.66

Zero retentionNo trainingCaching

Novita

100.00%

Context1M

Max output393.2K

p50 latency1,404 ms

Throughput236.0 t/s

Input price$0.44

Output price$1.32

No trainingCaching

Provider	Maximum context	Max output	p50 latency	p50 throughput	24h uptime	Input / 1M tokens	Output / 1M tokens
Deepinfra	1M	1M	466 ms	144.0 t/s	99.956%	$0.216	$0.647
Fireworks	1M	1M	1,491 ms	191.5 t/s	100.000%	$0.22	$0.66
Novita	1M	393.2K	1,404 ms	236.0 t/s	100.000%	$0.44	$1.32