modelbenchmark.io

DeepSeek-V3.1

DeepSeek · deepseek-v3-1

Compare

DeepSeek chat model for instruction following, coding, and analysis

Specification

most-agreed values

Context
128K
Max output
32K
Released
2025-08-19
Knowledge cutoff
2025-08
Retires
Open weights
no
Input
text
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.20
Output
$0.70
Cache read
$0.10
Cache write
$0.56

Available from 17 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
ollama · 671b-cloudfreefree164K164K
nano-gpt · deepseek-ai$0.20$0.70$0.10128K66K
nano-gpt · thinking$0.20$0.70$0.10128K66K
submodel · deepseek-ai$0.20$0.8075K164K
deepinfra · deepseek-ai$0.25$0.95$0.13164K8K
huggingface · deepseek-ai$0.27$1.00131K8K
meganova · deepseek-ai$0.27$1.00164K164K
siliconflow · deepseek-ai$0.27$1.00164K164K
baseten · deepseek-ai$0.50$1.50164K131K
wandb · deepseek-ai$0.55$1.65$0.55161K161K
wandb · wandb$0.55$1.65161K128K
pioneer · deepseek-ai$0.56$1.68$0.56$0.56164K131K
alibaba-cn$0.574$1.721131K66K
togetherai · deepseek-ai$0.60$1.70131K131K
replicate · replicate$0.672$2.016164K164K
sambanova · sambanova$3.00$4.50131K131K
qiniu-ai128K32K

Other listings of this model

same model, different key

More from DeepSeek

most-hosted first