---
title: "YandexGPT (language model) (NL)"
source: "https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)"
wiki: "systems-analysis.info/int"
article: "YandexGPT_(language_model)_(NL)"
language: "nl"
categories:
  - "Category:Dutch"
  - "Category:Large language models"
  - "Category:LLM families"
  - "Category:Machine learning"
revision_id: 8522
wiki_created_at: 2026-09-07T01:18:36Z
wiki_modified_at: 2026-09-07T01:18:36Z
downloaded_at: 2026-09-07T23:25:50Z
---

# YandexGPT (language model) (NL)

**YandexGPT (Yet another GPT)** — een familie van grote taalmodellen (Large Language Models), ontwikkeld door Yandex en voor het eerst gepresenteerd in mei 2023.<sup>[\[1\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-rbc-may17-1)</sup> De YandexGPT-netwerken worden gebruikt in de ingebouwde assistent Alice, in de Zoekmachine en andere diensten, en zijn ook beschikbaar via de publieke API van het Yandex Cloud-platform.<sup>[\[2\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-cloud-api-2)</sup>

**YaLM‑100B** (2022) — het voorafgaande onderzoeksmodel met 100 miljard parameters en open broncode; het diende als "proof of concept", maar YandexGPT werd afzonderlijk ontwikkeld voor commercieel gebruik.<sup>[\[3\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-github-yalm-3)</sup>

## Geschiedenis van releases

| Datum       | Release                   | Kernkenmerken                                                                                                                                                                  |
|-------------|---------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Jun 2022    | YaLM‑100B                 | 100 B parameters, 1,7 TB data; Apache 2.0.<sup>[\[3\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-github-yalm-3)</sup>                        |
| 17 mei 2023 | YandexGPT 1.0             | Integratie in "Alice".<sup>[\[1\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-rbc-may17-1)</sup>                                              |
| 7 sep 2023  | YandexGPT 2               | +67 % kwaliteit volgens interne tests.<sup>[\[4\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-rbc-sep8-4)</sup>                               |
| 28 mrt 2024 | YandexGPT 3 Pro / Lite    | Nieuwe zakelijke API-lijn.<sup>[\[5\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-rbc-mar28-5)</sup>                                          |
| 24 okt 2024 | YandexGPT 4 Pro / Lite    | Context van 32.000 tokens; verborgen redenering (chain‑of‑thought).<sup>[\[6\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-habr-gpt4-6)</sup> |
| 25 feb 2025 | YandexGPT 5 Pro           | Pariteit met GPT‑4o in 64 % van de taken.<sup>[\[7\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-adindex-5pro-7)</sup>                        |
| 31 mrt 2025 | YandexGPT 5 Lite Instruct | Model van 8 miljard parameters in open toegang; Llama-formaat.<sup>[\[8\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-hf-5lite-8)</sup>       |

Belangrijkste versies

## Architectuur en training

- Basisarchitectuur: transformer, geoptimaliseerd voor de Russische taal.
- YandexGPT 5 Lite: Llama-compatibel; pre-training ≈ 15 biljoen tokens, aansluitende fine-tuning ≈ 320 miljard.<sup>[\[8\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-hf-5lite-8)</sup>

### Context en limieten

- Architecturaal maximale contextlengte — 32.000 tokens (versies 4/5).<sup>[\[6\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-habr-gpt4-6)</sup>
- De publieke API beperkt één verzoek (prompt + completion) tot 7.400 tokens.<sup>[\[9\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-langchain-7400-9)</sup>
- Maximale grootte van het \*\*antwoord\*\* — 2.000 tokens volgens de sectie "Quotas and limits".<sup>[\[10\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-cloud-limits-10)</sup>

## Huidige modellen (juni 2025)

| Model            | Parameters | Context | Licentie                | Opmerkingen                                                                                                                                        |
|------------------|------------|---------|-------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------|
| YandexGPT 5 Pro  | n.v.t.     | 32.000  | proprietair             | Toegankelijk via API en "Alice Pro".<sup>[\[7\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-adindex-5pro-7)</sup> |
| YandexGPT 5 Lite | 8 mld      | 32.000  | Yandex GPT‑Lite License | Open; Llama-compatibel.<sup>[\[8\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-hf-5lite-8)</sup>                  |
| YaLM‑100B        | 100 mld    | 2.048   | Apache 2.0              | Oorspronkelijk project.<sup>[\[3\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-github-yalm-3)</sup>               |

## Benchmarks

- Interne tests: 5 Pro bereikte pariteit met GPT‑4o in 64 % van de taken; overschrijding ten opzichte van 4 Pro — 67 %.<sup>[\[7\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-adindex-5pro-7)</sup>
- ru‑LLM Arena: YandexGPT handhaaft de leiderspositie op basis van ELO-rating onder Russischtalige modellen.<sup>[\[11\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-llmarena-11)</sup>

## Fine‑tuning

Voor 5 Lite wordt de LoRA-methode officieel ondersteund; een voorbeeld van uitvoering is gepubliceerd in de model card.<sup>[\[8\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-hf-5lite-8)</sup>

## API‑modi

- Synchroon — snelle antwoorden (Lite).
- Asynchroon — resource-intensieve taken (Pro).<sup>[\[2\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-cloud-api-2)</sup>

## Multimodaliteit

De YandexGPT-familie blijft tekstgebaseerd; multimodale diensten ("Neuro", "YandexArt", "Yandex Vision") worden afzonderlijk ontwikkeld.<sup>[\[6\]](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_note-habr-gpt4-6)</sup>

## Verwijzingen

- YandexGPT in Yandex Cloud — servicepagina
- Gewichten van YandexGPT‑5 Lite op Hugging Face
- YaLM‑100B-repository op GitHub

## Literatuur

- Matkin, N. et al. (2024). *Comparative Analysis of Encoder-Based NER and Large Language Models for Skill Extraction from Russian Job Vacancies*. arXiv:2407.19816.
- Tsanda, A.; Bruches, E. (2024). *Russian-Language Multimodal Dataset for Automatic Summarization of Scientific Papers*. arXiv:2405.07886.
- Goloburda, M. et al. (2025). *Qorǵau: Evaluating LLM Safety in Kazakh-Russian Bilingual Contexts*. arXiv:2502.13640.
- Togmanov, M. et al. (2025). *KazMMLU: Evaluating Language Models on Kazakh, Russian, and Regional Knowledge of Kazakhstan*. arXiv:2502.12829.
- Noels, S. et al. (2025). *What Large Language Models Do Not Talk About: An Empirical Study of Moderation and Censorship Practices*. arXiv:2504.03803.

## Noten

1.  <span id="cite_note-rbc-may17-1">↑ <sup>[1.0](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-rbc-may17_1-0)</sup> <sup>[1.1](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-rbc-may17_1-1)</sup> «“Яндекс” добавил в “Алису” аналог ChatGPT». *РБК*. <a href="https://www.rbc.ru/technology_and_media/17/05/2023/6464aa4e9a79473a1f0464c1" class="external autonumber" rel="nofollow">[1]</a></span>
2.  <span id="cite_note-cloud-api-2">↑ <sup>[2.0](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-cloud-api_2-0)</sup> <sup>[2.1](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-cloud-api_2-1)</sup> «Getting started with YandexGPT (Quickstart)». *Yandex Cloud Docs*. <a href="https://cloud.yandex.com/en/docs/yandexgpt/quickstart" class="external autonumber" rel="nofollow">[2]</a></span>
3.  <span id="cite_note-github-yalm-3">↑ <sup>[3.0](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-github-yalm_3-0)</sup> <sup>[3.1](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-github-yalm_3-1)</sup> <sup>[3.2](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-github-yalm_3-2)</sup> «yandex/YaLM‑100B: Pretrained language model with 100B». *GitHub*. <a href="https://github.com/yandex/YaLM-100B" class="external autonumber" rel="nofollow">[3]</a></span>
4.  <span id="cite_note-rbc-sep8-4">[↑](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-rbc-sep8_4-0) «Как “Яндекс” решил зарабатывать на своём аналоге ChatGPT». *РБК*. <a href="https://www.rbc.ru/technology_and_media/08/09/2023/64f9ff099a79472ad0e24715" class="external autonumber" rel="nofollow">[4]</a></span>
5.  <span id="cite_note-rbc-mar28-5">[↑](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-rbc-mar28_5-0) «“Яндекс” представил третье поколение нейросетей YandexGPT». *РБК*. <a href="https://www.rbc.ru/rbcfreenews/660563e29a794766b508c1f9" class="external autonumber" rel="nofollow">[5]</a></span>
6.  <span id="cite_note-habr-gpt4-6">↑ <sup>[6.0](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-habr-gpt4_6-0)</sup> <sup>[6.1](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-habr-gpt4_6-1)</sup> <sup>[6.2](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-habr-gpt4_6-2)</sup> «Более мощное семейство моделей YandexGPT 4». *Habr*. <a href="https://habr.com/ru/companies/yandex/articles/852968/" class="external autonumber" rel="nofollow">[6]</a></span>
7.  <span id="cite_note-adindex-5pro-7">↑ <sup>[7.0](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-adindex-5pro_7-0)</sup> <sup>[7.1](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-adindex-5pro_7-1)</sup> <sup>[7.2](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-adindex-5pro_7-2)</sup> «“Яндекс” внедрил YandexGPT 5 Pro в чат с “Алисой Про”». *AdIndex*. <a href="https://adindex.ru/news/digital/2025/02/25/331154.phtml" class="external autonumber" rel="nofollow">[7]</a></span>
8.  <span id="cite_note-hf-5lite-8">↑ <sup>[8.0](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-hf-5lite_8-0)</sup> <sup>[8.1](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-hf-5lite_8-1)</sup> <sup>[8.2](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-hf-5lite_8-2)</sup> <sup>[8.3](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-hf-5lite_8-3)</sup> «yandex/YandexGPT‑5‑Lite‑8B‑pretrain». *Hugging Face*. <a href="https://huggingface.co/yandex/YandexGPT-5-Lite-8B-pretrain" class="external autonumber" rel="nofollow">[8]</a></span>
9.  <span id="cite_note-langchain-7400-9">[↑](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-langchain-7400_9-0) «ChatYandexGPT API Reference (max_tokens = 7400)». *LangChain Docs*. <a href="https://python.langchain.com/api_reference/community/chat_models/langchain_community.chat_models.yandex.ChatYandexGPT.html" class="external autonumber" rel="nofollow">[9]</a></span>
10. <span id="cite_note-cloud-limits-10">[↑](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-cloud-limits_10-0) «Yandex Cloud service quotas and limits → Foundation Models». *Yandex Cloud Docs*. <a href="https://cloud.yandex.com/en/docs/overview/concepts/quotas-limits" class="external autonumber" rel="nofollow">[10]</a></span>
11. <span id="cite_note-llmarena-11">[↑](https://systems-analysis.info/int/YandexGPT_(language_model)_(NL)#cite_ref-llmarena_11-0) «llmarena/llmarena — российская краудсорсинговая платформа оценки LLM». *GitHub*. <a href="https://github.com/llmarena/llmarena" class="external autonumber" rel="nofollow">[11]</a></span>