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)
87 provisional overall
Avg. top 3 provisional
65
Provisional-ranked
8
Verified-ranked
1
Provider models
Browse all models| Model | Status | Context | Price | Speed | TTFT | Provisional score |
|---|---|---|---|---|---|---|
| DeepSeek V4 Pro (Max) | Current | 1M | $1.74 / $3.48 | N/A | N/A | 87Verified-ranked |
| DeepSeek V3.2 | Established | 128K | $0.28 / $0.42 | 35 | 3.75s | ~57 |
| DeepSeek LLM 2.0 | Tracked | 128K | $0.00 / $0.00 | N/A | N/A | ~51 |
| DeepSeek Coder 2.0 | Tracked | 128K | N/A | N/A | N/A | ~51 |
| DeepSeekMath V2 | Tracked | 128K | $0.00 / $0.00 | N/A | N/A | ~50 |
| DeepSeek V3 | Established | 128K | $0.27 / $1.10 | N/A | N/A | ~35 |
| DeepSeek-R1 | Established | 128K | $0.55 / $2.19 | N/A | N/A | ~33 |
| DeepSeek V3.1 | Established | 128K | $0.00 / $0.00 | N/A | N/A | ~25 |
| DeepSeek R1 Distill Qwen 32B | Established | 128K | $0.00 / $0.00 | 60 | 0.84s | — |