Title: DeepSeek-V3.2 %E2%80%94 deepseek-v3-2
Canonical: https://llmeu.com/sk/models/deepseek-v3-2/
Markdown mirror: https://llmeu.com/sk/models/deepseek-v3-2.md
Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json
Provider: DeepSeek
Licence: mit
Hosting: catalogue only %E2%80%94 not hosted on LLM EU yet
Sources: https://huggingface.co/deepseek-ai/DeepSeek-V3.2

# DeepSeek-V3.2

[DeepSeek](/sk/providers/deepseek/)

Naposledy skontrolovan%C3%A9: 14. 10. 2026

## Suverenita a hosting

Tento model je v katal%C3%B3gu len pre referenciu. Zatia%C4%BE nie je hostovan%C3%BD na LLM EU a %C5%BEiadny koncov%C3%BD bod prvej strany ho tu neobsluhuje. Po%C5%BEiadavky na%C5%88 smeruj%C3%BA priamo k poskytovate%C4%BEovi, pod%C4%BEa jeho vlastn%C3%BDch podmienok.

## Zhrnutie

DeepSeek-V3.2 je model typu mixture-of-experts s 685 miliardami parametrov, vydan%C3%BD pod licenciou MIT, s konfigura%C4%8Dn%C3%BDm maximom 163 840 tokenov. Jeho hlavnou architektonickou zmenou je DeepSeek Sparse Attention, ktor%C3%BA dod%C3%A1vate%C4%BE opisuje ako zn%C3%AD%C5%BEenie v%C3%BDpo%C4%8Dtov%C3%BDch n%C3%A1kladov na inferenciu s dlh%C3%BDm kontextom. Je to model iba pre text a vy%C5%BEaduje nasadenie na viacer%C3%BDch uzloch s plnou presnos%C5%A5ou.

## %C5%A0pecifik%C3%A1cie

- Poskytovate%C4%BE
  DeepSeek (CN)
- Kontextov%C3%A9 okno
  164k
- Parametre
  685.36B
- Modalit%C3%A1
  text
- Licencia
  mit %C2%B7 [licence](https://opensource.org/license/mit)
- Otvoren%C3%A9 v%C3%A1hy
  yes
- D%C3%A1tum vydania
  01. 12. 2025
- %C3%9Alohy
  [Uva%C5%BEovanie](/sk/tasks/reasoning/), [K%C3%B3dovanie](/sk/tasks/coding/), [Zhrnutie](/sk/tasks/summarization/), [Vyh%C4%BEad%C3%A1vanie (RAG)](/sk/tasks/rag/)

## Jazyky

**Siln%C3%A1:** en, zh

**Dostato%C4%8Dn%C3%A1:** de, fr, es, it, pt, ru, ja, ko, ar

#### Siln%C3%A1

- en
- zh

#### Dostato%C4%8Dn%C3%A1

- de
- fr
- es
- it
- pt
- ru
- ja
- ko
- ar

Neoveren%C3%A9: poskytovate%C4%BE tento %C3%BAdaj nedokumentoval. %E2%80%94 a language is listed as strong only where a source claims it; the card links the evidence where one exists.

## Kedy ho pou%C5%BEi%C5%A5

#### Siln%C3%A9 str%C3%A1nky

- MIT licence
- sparse attention aimed at lower long-context inference cost
- 163,840-token configuration maximum

#### Obmedzenia

- 685B parameters: full-precision serving needs several nodes
- vendor efficiency claims are not independently reproduced here
- no evaluation has been run by LLM EU

## %C4%8Casto kladen%C3%A9 ot%C3%A1zky

### Is DeepSeek-V3.2 served on LLM EU?

It has no endpoint in the catalogue, so it is catalogue only %E2%80%94 not hosted on LLM EU yet, with no region and no retention to report. A card whose status is "listed" is a catalogue entry, not a running deployment.

### Under what licence is DeepSeek-V3.2 released?

DeepSeek-V3.2 is recorded under mit. The recorded licence text is https://opensource.org/license/mit. The weights are downloadable (open_weights is true). LLM EU republishes the licence id and the link and gives no legal advice; the licence text governs.

### How large is DeepSeek-V3.2, and what context window does it have?

DeepSeek-V3.2 is recorded at 685B (safetensors total 685,355,329,792) (source: https://huggingface.co/api/models/deepseek-ai/DeepSeek-V3.2), with a context window of 163840 tokens (source: https://huggingface.co/deepseek-ai/DeepSeek-V3.2/blob/main/config.json).

### Ceny

Tento model je v katal%C3%B3gu len pre referenciu. Zatia%C4%BE nie je hostovan%C3%BD na LLM EU a %C5%BEiadny koncov%C3%BD bod prvej strany ho tu neobsluhuje. Po%C5%BEiadavky na%C5%88 smeruj%C3%BA priamo k poskytovate%C4%BEovi, pod%C4%BEa jeho vlastn%C3%BDch podmienok.

### Stav overenia

Ak pole nem%C3%A1 zdroj, karta ho zobraz%C3%AD ako neoveren%C3%A9, namiesto toho, aby ho vyplnila. Pr%C3%A1zdne pole je inform%C3%A1cia.

LLM EU Verified znamen%C3%A1 zdokumentovan%C3%A9, nie pr%C3%A1vne certifikovan%C3%A9. Nie je to pos%C3%BAdenie zhody pod%C4%BEa AI Act a LLM EU nie je org%C3%A1n E%C3%9A.

### Verzie a varianty

- catalogue %C2%B7 Iba v katal%C3%B3gu %C2%B7 [weights](https://huggingface.co/deepseek-ai/DeepSeek-V3.2)

### Pr%C3%ADklad po%C5%BEiadavky

```
curl https://api.llmeu.com/v1/chat/completions \
  -H "Authorization: Bearer $LLMEU_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v3-2","messages":[{"role":"user","content":"Hello"}]}'
```

Pre tento model nie je nasaden%C3%BD %C5%BEiadny playground: neobsluhuje ho %C5%BEiadny koncov%C3%BD bod, tak%C5%BEe po%C5%BEiadavka by nemala kam %C3%ADs%C5%A5.

### Benchmarky

%C5%BDiadne zverejnen%C3%A9 hodnotenie. Toto je medzera, nie skryt%C3%BD v%C3%BDsledok.

### Zdroje

- [huggingface.co](https://huggingface.co/deepseek-ai/DeepSeek-V3.2)

### Zmeny karty

- 01. 12. 2025 %E2%80%94 D%C3%A1tum vydania
- 14. 10. 2026 %E2%80%94 Naposledy skontrolovan%C3%A9

Naposledy skontrolovan%C3%A9: 14. 10. 2026