← All models

DeepSeek V4.1 Flash

deepseek-v4p1-flash

Multimodal open-weight MoE model for long-context reasoning and agentic work

Intelligence
Pending
Context
1.04M
Added
2026-09-25
Inputs
Vision

Providers

Fireworks AI primary

accounts/fireworks/models/deepseek-v4p1-flash

Uptime 92.1%
Input / 1M
$0.30
Output / 1M
$1.20
TTFT
0.35s
Output rate
15.4 tok/s
131,072 max output cache $0.006 read / $0.30 write per 1M ZDR Training No Provider terms
Prompt caching Automatic
Together AI failover 1

deepseek-ai/DeepSeek-V4.1-Flash

Uptime 100.0%
Input / 1M
$0.30
Output / 1M
$1.20
TTFT
2.66s
Output rate
287.9 tok/s
131,072 max output cache $0.006 read / $0.30 write per 1M ZDR Training No Provider terms
Prompt caching Automatic
Inco failover 2

deepseek-v4.1-flash:fast

Uptime 100.0%
Input / 1M
$0.60
Output / 1M
$2.40
TTFT
0.57s
Output rate
444.0 tok/s
131,072 max output cache $0.012 read / $0.60 write per 1M Retention Yes, period not published Training No Provider terms
Prompt caching Automatic