DeepSeek
DeepSeek V3.1 (Non-reasoning)
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...
DeepSeek release notesModel Snapshot
ReleaseAug 21, 2025
Context163,840
Overall
Rank #212 across 566
21.0
Coding
No rank
n/a
Blended Price
Rank #158 across 371
$0.840/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Max Output32,768
Streaming speed is not measured for this model yet.
Pricing
Blended$0.840/M
Input$0.560/M
Output$1.68/M
Metadata
Modalities
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 21.0 | #212 |
| Artificial Analysis Math Index | math | 49.7 | #141 |
| MMLU-Pro | reasoning | 83.3% | #60 |
| GPQA | reasoning | 73.5% | #220 |
| Humanity's Last Exam |