Model card
DeepSeek-V3.2.
DeepSeekopen-source
§ 02 · Benchmarks
Every benchmark DeepSeek-V3.2 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | AIME 2025 | Reasoning · Mathematical Reasoning | accuracy | 93.1% | #6 | — | source ↗ |
| 02 | Tau2-Bench | Agentic AI · Tool Use | accuracy | 80.3% | #6 | — | source ↗ |
| 03 | LiveCodeBench | Computer Code · Code Generation | pass-1 | 83.3% | #8 | — | source ↗ |
| 04 | SWE-Bench Verified | Computer Code · Code Generation | accuracy | 73.1% | #12 | — | source ↗ |
| 05 | BrowseComp | Natural Language Processing · Question Answering | accuracy | 51.4% | #13 | — | source ↗ |
| 06 | HLE | Reasoning · Multi-step Reasoning | accuracy | 25.1% | #25 | — | source ↗ |
| 07 | GPQA Diamond | Reasoning · Multi-step Reasoning | accuracy | 82.4% | #31 | — | source ↗ |
| 08 | PLCC | Natural Language Processing · Polish Cultural Competency | culture-and-tradition | 78.0% | #41 | — | source ↗ |
| 09 | PLCC | Natural Language Processing · Polish Cultural Competency | history | 82.0% | #53 | — | source ↗ |
| 10 | PLCC | Natural Language Processing · Polish Cultural Competency | vocabulary | 65.0% | #53 | — | source ↗ |
| 11 | PLCC | Natural Language Processing · Polish Cultural Competency | average | 71.7% | #54 | — | source ↗ |
| 12 | PLCC | Natural Language Processing · Polish Cultural Competency | art-and-entertainment | 61.0% | #61 | — | source ↗ |
| 13 | PLCC | Natural Language Processing · Polish Cultural Competency | grammar | 66.0% | #61 | — | source ↗ |
| 14 | PLCC | Natural Language Processing · Polish Cultural Competency | geography | 78.0% | #68 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where DeepSeek-V3.2 actually performs.
§ 04 · Papers
1 paper with results for DeepSeek-V3.2.
- 2025-12-02· 7 results
DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
§ 05 · Related models
Other DeepSeek models scored on Codesota.
§ 06 · Sources & freshness
Where these numbers come from.
pwc-dump
7
results
sdadas/PLCC
7
results
7 of 14 rows marked verified.