DeepSeek
DeepSeek · China · Language & multimodal
China Open weightsIndependent test
Showing version: DeepSeek-V3.2 · 2025-12-01 · Version lineage ↓
01
Overview
DeepSeek is a family of large language models known for MoE architecture and open weights, with the V3 line for general use and the R1 line for reasoning.
Context length163,840 tokensInput$0.26 USD / 1M tokensRelease date2025-12-01
02 / Provenance
Field checks and source notes
When sources disagree on a field, every reading is listed here with its own citation. We do not take a midpoint, invent a missing figure, or decide which side is right — follow the citation and judge for your case.
These field-level notes are recorded verbatim in Chinese in the source data. Open the Chinese version
03 / Plain layer
Plain-language benchmarks
Coding 70%mini-SWE-agent (bash only), high setting, run date 2026-02-17 as listed on source page
Memory 163,840 tokens
Model size Not disclosed
Price / free $0.26 / $0.38 USD / 1M tokens
04
Practical facts
| Context length | 163,840 tokens |
|---|---|
| Max output | Not disclosed |
| Parameters | Not disclosed |
| Input | $0.26 USD / 1M tokens |
| Output price | $0.38 USD / 1M tokens |
| CNY | Not disclosed |
| Free tier | Not disclosed |
| Output speed | Not disclosed |
| Release date | |
| License | Not disclosed |
Show technical data
Technical benchmark table
This model is evaluated per reasoning tier; available tiers: non-reasoning / reasoning. Each row below shows the highest tier measured for that benchmark. Tier switching happens on the compare page.
| Benchmark | Score | Max / unit | Source type | Source | Test date | Conditions |
|---|---|---|---|---|---|---|
| SWE-bench Verified swe-bench-verified | 70% | percent / 100 | Independent test | Tierhigh |
| Version | Release date | Status | Openness | Parameters | What changed | Detail | |
|---|---|---|---|---|---|---|---|
| DeepSeek-V3.2Showing version | Active | Open weights | Not disclosed | TBD | Detail | ||
| DeepSeek-V3.2-Exp | Active | Open weights | Not disclosed | TBD | Detail | ||
| DeepSeek-V3.1 | Active | Open weights | Not disclosed | TBD | Detail | ||
| DeepSeek-R1 | Active | Open weights | Not disclosed | TBD | Detail | ||
| DeepSeek-V3 | Active | Open weights | Not disclosed | TBD | Detail | ||
| DeepSeek-V2 | Active | Open weights | Not disclosed | TBD | Detail |
Last updated: · Sources on detail page
06 / Compare