Skip to main content
Provider profile

DeepSeek

BenchLM aggregates canonical model families under DeepSeek so you can compare release depth and model quality at the provider level instead of jumping across single model pages. Scores shown here are provisional unless otherwise noted.

Top model

DeepSeek V4 Pro (Max)

88 provisional overall

Avg. top 3 provisional

66

Provisional-ranked

8

Verified-ranked

1

Provider models

Browse all models
ModelStatusContext Price Speed TTFTProvisional score
DeepSeek V4 Pro (Max)Current1M$1.74 / $3.48N/AN/A88Verified-ranked
DeepSeek V3.2Current128K$0.28 / $0.42353.75s~58
DeepSeek Coder 2.0Tracked128KN/AN/AN/A~52
DeepSeek LLM 2.0Tracked128K$0.00 / $0.00N/AN/A~51
DeepSeekMath V2Tracked128K$0.00 / $0.00N/AN/A~50
DeepSeek V3Established128K$0.27 / $1.10N/AN/A~36
DeepSeek-R1Established128K$0.55 / $2.19N/AN/A~33
DeepSeek V3.1Established128K$0.00 / $0.00N/AN/A~26
DeepSeek R1 Distill Qwen 32BEstablished128K$0.00 / $0.00600.84s