Title: Gemma 4 E4B IT %E2%80%94 gemma-4-e4b-it
Canonical: https://llmeu.com/lt/models/gemma-4-e4b-it/
Markdown mirror: https://llmeu.com/lt/models/gemma-4-e4b-it.md
Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json
Provider: Google DeepMind
Licence: apache-2.0
Hosting: catalogue only %E2%80%94 not hosted on LLM EU yet
Sources: https://huggingface.co/google/gemma-4-E4B-it

# Gemma 4 E4B IT

[Google DeepMind](/lt/providers/google/)

Paskutin%C4%AF kart%C4%85 per%C5%BEi%C5%ABr%C4%97ta: 2026-10-14

## Suverenitetas ir talpinimas

%C5%A0is modelis yra kataloge tik nuorodai. Jis dar n%C4%97ra talpinamas LLM EU, ir joks pirmosios %C5%A1alies galinis ta%C5%A1kas jo neaptarnauja. U%C5%BEklausos jam siun%C4%8Diamos tiesiogiai teik%C4%97jui, pagal to teik%C4%97jo s%C4%85lygas.

## Santrauka

Gemma 4 E4B IT yra ma%C5%BEiausias Gemma 4 modelis su nat%C5%ABraliu garsu, apdorojantis tekst%C4%85, vaizdus ir garso %C4%AFvest%C4%AF su 128K %C5%BEeton%C5%B3 konteksto langu. Google teigia, kad %E2%80%9EE%E2%80%9C raid%C4%97 E2B ir E4B rei%C5%A1kia efektyvius parametrus ir kad %C5%A1ie dyd%C5%BEiai naudoja sluoksnio %C4%AFterpimus (Per-Layer Embeddings), tod%C4%97l saugomas modelis yra apie 8 mlrd. parametr%C5%B3, o efektyvus skai%C4%8Dius yra ma%C5%BEesnis. Modelio kortel%C4%97je nurodoma 69,4% MMLU Pro ir 58,6% GPQA Diamond, ir jis yra licencijuotas pagal Apache-2.0.

## Specifikacijos

- Teik%C4%97jas
  Google DeepMind (US)
- Konteksto langas
  131k
- Parametrai
  8B
- Modalumai
  text, vision, audio
- Licencija
  apache-2.0 %C2%B7 [licence](https://ai.google.dev/gemma/docs/gemma_4_license)
- Atvirieji svoriai
  yes
- I%C5%A1leidimo data
  2026-03-02
- U%C5%BEduotys
  [Ma%C5%BEa kaina](/lt/tasks/cheap/), [Vizija](/lt/tasks/vision/), [Daugiakalbis palaikymas](/lt/tasks/multilingual-support/), [Apibendrinimas](/lt/tasks/summarization/)

## Kalbos

**Stipri:** en, de, fr, es, it, pt

**Pakankama:** nl, pl, sv, cs, ro, el, uk, ru, ar, hi, zh, ja, ko, tr

#### Stipri

- en
- de
- fr
- es
- it
- pt

#### Pakankama

- nl
- pl
- sv
- cs
- ro
- el
- uk
- ru
- ar
- hi
- zh
- ja
- ko
- tr

Nepatvirtinta: teik%C4%97jas nepagrind%C4%97 %C5%A1io teiginio. %E2%80%94 a language is listed as strong only where a source claims it; the card links the evidence where one exists.

## Kada naudoti

#### Stipryb%C4%97s

- native audio input in the small-model class
- Apache-2.0 licence
- runs on modest hardware
- text, image and audio in one checkpoint

#### Apribojimai

- reasoning scores well below the larger Gemma 4 sizes
- 128K-token window, half of the medium models
- no evaluation has been run by LLM EU

## Da%C5%BEniausiai u%C5%BEduodami klausimai

### Is Gemma 4 E4B IT 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 Gemma 4 E4B IT released?

Gemma 4 E4B IT is recorded under apache-2.0. The recorded licence text is https://ai.google.dev/gemma/docs/gemma_4_license. 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 Gemma 4 E4B IT, and what context window does it have?

Gemma 4 E4B IT is recorded at about 8B stored (safetensors total 7,996,156,490); the card states that "E" in E2B and E4B stands for effective parameters, using Per-Layer Embeddings (source: https://huggingface.co/api/models/google/gemma-4-E4B-it), with a context window of 131072 tokens (source: https://huggingface.co/google/gemma-4-E4B-it).

### Kainodara

%C5%A0is modelis yra kataloge tik nuorodai. Jis dar n%C4%97ra talpinamas LLM EU, ir joks pirmosios %C5%A1alies galinis ta%C5%A1kas jo neaptarnauja. U%C5%BEklausos jam siun%C4%8Diamos tiesiogiai teik%C4%97jui, pagal to teik%C4%97jo s%C4%85lygas.

### Patvirtinimo b%C5%ABsena

Kai laukas neturi %C5%A1altinio, kortel%C4%97je jis rodomas kaip nepatikrintas, o ne u%C5%BEpildomas. Tu%C5%A1%C4%8Dias laukas yra informacija.

LLM EU Verified rei%C5%A1kia dokumentuota, o ne teisi%C5%A1kai sertifikuota. Tai n%C4%97ra AI akto atitikties vertinimas, ir LLM EU n%C4%97ra ES institucija.

### Versijos ir variantai

- catalogue %C2%B7 Tik kataloge %C2%B7 [weights](https://huggingface.co/google/gemma-4-E4B-it)

### U%C5%BEklausos pavyzdys

```
curl https://api.llmeu.com/v1/chat/completions \
  -H "Authorization: Bearer $LLMEU_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemma-4-e4b-it","messages":[{"role":"user","content":"Hello"}]}'
```

%C5%A0iam modeliui n%C4%97ra %C4%AFdiegta bandym%C5%B3 aik%C5%A1tel%C4%97: joks galinis ta%C5%A1kas jo neaptarnauja, tod%C4%97l u%C5%BEklausa netur%C4%97t%C5%B3 kur nueiti.

### Etaloniniai testai

- MMLU Pro: **69.4% (vendor-reported)** [source](https://huggingface.co/google/gemma-4-E4B-it)
- AIME 2026, no tools: **42.5% (vendor-reported)** [source](https://huggingface.co/google/gemma-4-E4B-it)
- GPQA Diamond: **58.6% (vendor-reported)** [source](https://huggingface.co/google/gemma-4-E4B-it)

### %C5%A0altiniai

- [huggingface.co](https://huggingface.co/google/gemma-4-E4B-it)

### Kortel%C4%97s pakeitim%C5%B3 %C5%BEurnalas

- 2026-03-02 %E2%80%94 I%C5%A1leidimo data
- 2026-10-14 %E2%80%94 Paskutin%C4%AF kart%C4%85 per%C5%BEi%C5%ABr%C4%97ta

Paskutin%C4%AF kart%C4%85 per%C5%BEi%C5%ABr%C4%97ta: 2026-10-14

### U%C5%BEduotys

- [Ma%C5%BEa kaina](/lt/tasks/cheap/)