Title: gpt-oss-20b | LLM EU
Canonical: https://llmeu.com/mt/models/gpt-oss-20b/
Markdown mirror: https://llmeu.com/mt/models/gpt-oss-20b.md
Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json
Provider: OpenAI
Licence: apache-2.0
Hosting: catalogue only %E2%80%94 not hosted on LLM EU yet
Sources: https://huggingface.co/openai/gpt-oss-20b

# gpt-oss-20b

[OpenAI](/mt/providers/openai/)

A%C4%A7%C4%A7ar revi%C5%BCjoni: 14 Ott 2026

## Sovranit%C3%A0 u ospitar

Dan il-mudell huwa kkatalgat g%C4%A7al referenza. G%C4%A7adu mhux ospitat fuq LLM EU, u l-ebda endpoint tal-ewwel parti ma jservih hawn. It-talbiet lejh imorru direttament g%C4%A7and il-fornitur, ta%C4%A7t it-termini tal-fornitur stess.

## Sommarju

gpt-oss-20b huwa l-i%C5%BCg%C4%A7ar mi%C5%BC-%C5%BCew%C4%A1 mudelli ta' ra%C4%A1unament b'pi%C5%BC miftu%C4%A7 ta' OpenAI: 21B parametru totali, 3.6B attivi, li%C4%8Benzjat Apache-2.0 u %C5%BCg%C4%A7ir bi%C5%BC%C5%BCejjed biex ja%C4%A7dem f'madwar 16GB ta' memorja. Il-konfigurazzjoni tieg%C4%A7u tiddikjara l-istess tieqa ta' kuntest ta' 131,072 token b%C4%A7all-mudell 120B. Jimmira skjeramenti lokali u sensittivi g%C4%A7al latenza aktar milli kwalit%C3%A0 massima.

## Spe%C4%8Bifikazzjonijiet

- Fornitur
  OpenAI (US)
- Tieqa tal-kuntest
  131k
- Parametri
  20.91B
- Modalitajiet
  text
- Li%C4%8Benzja
  apache-2.0 %C2%B7 [licence](https://www.apache.org/licenses/LICENSE-2.0)
- Pi%C5%BCijiet miftu%C4%A7a
  yes
- Data tar-rilaxx
  05 Aww 2025
- Kompiti
  [Ra%C4%A1unament](/mt/tasks/reasoning/), [Spi%C5%BCa baxxa](/mt/tasks/cheap/), [Ikkodifikar](/mt/tasks/coding/), [Irkupru (RAG)](/mt/tasks/rag/)

## Lingwi

**Qawwi:** en

**Adekwat:** de, fr, es, it, pt, ja, zh

#### Qawwi

- en

#### Adekwat

- de
- fr
- es
- it
- pt
- ja
- zh

Mhux verifikat: il-fornitur ma ddokumentax din id-dikjarazzjoni. %E2%80%94 a language is listed as strong only where a source claims it; the card links the evidence where one exists.

## Meta tu%C5%BCah

#### Sa%C4%A7%C4%A7iet

- runs in about 16GB of memory
- same 131,072-token window as the larger sibling
- Apache-2.0 licence
- reasoning-effort control

#### Limiti

- quality clearly below the 120B model
- harmony response format required
- no evaluation has been run by LLM EU

## Mistoqsijiet frequenti

### Is gpt-oss-20b 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 gpt-oss-20b released?

gpt-oss-20b is recorded under apache-2.0. The recorded licence text is https://www.apache.org/licenses/LICENSE-2.0. 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 gpt-oss-20b, and what context window does it have?

gpt-oss-20b is recorded at 21B total, 3.6B active (safetensors total 20,914,757,184) (source: https://huggingface.co/openai/gpt-oss-20b), with a context window of 131072 tokens (source: https://huggingface.co/openai/gpt-oss-20b/blob/main/config.json).

### Prezzijiet

Dan il-mudell huwa kkatalgat g%C4%A7al referenza. G%C4%A7adu mhux ospitat fuq LLM EU, u l-ebda endpoint tal-ewwel parti ma jservih hawn. It-talbiet lejh imorru direttament g%C4%A7and il-fornitur, ta%C4%A7t it-termini tal-fornitur stess.

### Status ta' verifika

Fejn qasam m'g%C4%A7andux sors, il-kard turih b%C4%A7ala mhux verifikat minflok ma timlih. Qasam vojt huwa informazzjoni.

LLM EU Verified ifisser dokumentat, mhux i%C4%8B%C4%8Bertifikat legalment. Mhuwiex valutazzjoni ta' konformit%C3%A0 mal-AI Act u LLM EU mhuwiex korp tal-UE.

### Ver%C5%BCjonijiet u varjanti

- catalogue %C2%B7 Katalgu biss %C2%B7 [weights](https://huggingface.co/openai/gpt-oss-20b)

### E%C5%BCempju ta' talba

```
curl https://api.llmeu.com/v1/chat/completions \
  -H "Authorization: Bearer $LLMEU_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-oss-20b","messages":[{"role":"user","content":"Hello"}]}'
```

L-ebda playground mhuwiex deplojat g%C4%A7al dan il-mudell: l-ebda endpoint ma jservih, allura rikjesta ma jkollha fejn tmur.

### Benchmarks

L-ebda evalwazzjoni ppubblikata. Din hija lakuna, mhux ri%C5%BCultat mo%C4%A7bi.

### Sorsi

- [huggingface.co](https://huggingface.co/openai/gpt-oss-20b)

### Re%C4%A1istru tal-bidliet tal-karta

- 05 Aww 2025 %E2%80%94 Data tar-rilaxx
- 14 Ott 2026 %E2%80%94 A%C4%A7%C4%A7ar revi%C5%BCjoni

A%C4%A7%C4%A7ar revi%C5%BCjoni: 14 Ott 2026