---
title: "O‘zbekistonda L4 ijarasi: dastlabki kirish | UZ Dynamics"
description: "NVIDIA L4: ariza orqali dastlabki kirish, mavjudlik va form-faktor aniqlashtirilmoqda. 24 GB GDDR6 · 0,3 TB/s · PCIe 4."
url: "https://uzdynamics.ai/uz/gpus/l4"
lang: "uz"
image: "https://uzdynamics.ai/uz/gpus/l4/opengraph-image/card?f08c4f789845417d"
alternates:
  ru: "https://uzdynamics.ai/ru/gpus/l4"
  uz: "https://uzdynamics.ai/uz/gpus/l4"
  en: "https://uzdynamics.ai/en/gpus/l4"
---

> For an index of this site's pages, see [llms.txt](https://uzdynamics.ai/llms.txt). A markdown copy of each page is at its address with `.md` appended.

# NVIDIA L4 ijarasi: dastlabki kirish

Ada Lovelace · PCIe (aniqlanmoqda)

Dastlabki kirish uchun arizalar qabul qilinmoqda. Quvvatlarni O‘zbekistonda joylashtirmoqdamiz. Narxlarni hozircha e’lon qilmaymiz. Mavjudligini javobda aniqlab beramiz.

L4

[L4 uchun ariza](https://uzdynamics.ai/uz/request?source=gpu-page%3Al4&gpu=l4)

Vazifangizga mos ijara shartlarini yuboramiz. Shartnoma tuzilguncha to‘lov talab qilinmaydi.

- **Har bir GPU xotirasi:** 24 GB GDDR6
- **Xotira o‘tkazuvchanligi:** 0,3 TB/s
- **Kartalar orasidagi aloqa:** PCIe 4, NVLink yo‘q

## NVIDIA L4 xususiyatlari

- **Arxitektura:** Ada Lovelace
- **E’lon qilingan yil:** 2023
- **Xotira:** 24 GB
- **Xotira turi:** GDDR6
- **Xotira o‘tkazuvchanligi:** 0,3 TB/s
- **FP32:** 30,3 TFLOPS
- **FP16/BF16 siyrakliksiz:** 121 TFLOPS
- **FP8 siyrakliksiz:** 242 TFLOPS
- **FP4 siyrakliksiz:** yo‘q
- **FP64:** NVIDIA karta spetsifikatsiyasida ko‘rsatilmagan
- **RT yadrolari:** ha
- **NVENC/NVDEC:** NVENC 2, NVDEC 4
- **MIG:** yo‘q
- **Kartalar orasidagi aloqa:** PCIe 4, NVLink yo‘q
- **Quvvat (TDP/TGP):** 72 Vt
- **Form-faktor:** PCIe (aniqlanmoqda)

Xotira har bir GPU uchun ko‘rsatilgan. Bir nechta GPU xotirasi alohida: katta model uchun mos dasturiy ta’minot va modelni kartalar orasida taqsimlash kerak. NVLink ularni bitta GPU qilmaydi.

L4 form-faktori aniqlashtirilmoqda. Xususiyatlar NVIDIA rasmiy materiallari bo‘yicha, 26-sentabr, 2026 sanasida tekshirilgan.

## L4 vazifalar bo‘yicha

- Inferens: 14 mlrd parametrgacha LLM (FP8), bitta kartada
- Qo‘shimcha o‘qitish: QLoRA 14 mlrd parametrgacha, LoRA 8 mlrd parametrgacha, bitta kartada
- Noldan o‘qitish: 1 mlrd parametrgacha model, bitta kartada
- Render: ko‘pi bilan 16 GB hajmdagi sahnalar
- FP64 hisob-kitoblari: bu karta mos emas

Ochiq formulalar asosidagi taxminiy hisob: bu amaliy sinov natijasi yoki hisob-faktura emas.

## NVIDIA L4 o‘rnatilgan server

Server konfiguratsiyasi (protsessor, xotira, disk, tarmoq) aniqlanmoqda.

Hisob farazi: bitta serverda 8 tagacha GPU. Haqiqiy konfiguratsiyani javobda tasdiqlaymiz.

[8 tadan ortiq GPU? Klasterni muhokama qilish](https://uzdynamics.ai/uz/enterprise?source=gpu-page-cluster%3Al4&gpu=l4&count=more)

## Manbalar

- [NVIDIA L4](https://www.nvidia.com/en-us/data-center/l4/)
- [NVIDIA L4 GPU Accelerator Product Brief](https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/l4/PB-11316-001_v01.pdf)
- [NVIDIA Ada GPU Architecture (v2.02)](https://images.nvidia.com/aem-dam/Solutions/geforce/ada/nvidia-ada-gpu-architecture.pdf)
- [NVIDIA Video Encode and Decode Support Matrix](https://developer.nvidia.com/video-encode-decode-support-matrix)
- [NVIDIA Multi-Instance GPU User Guide](https://docs.nvidia.com/datacenter/tesla/mig-user-guide/latest/supported-gpus.html)
- [NVIDIA Newsroom, March 21, 2023](https://nvidianews.nvidia.com/news/nvidia-launches-inference-platforms-for-large-language-models-and-generative-ai-workloads)

## O‘xshash GPU modellari

- [NVIDIA RTX A5000](https://uzdynamics.ai/uz/gpus/rtx-a5000)
  24 GB GDDR6 · 0,77 TB/s · NVLink ko‘prigi
  L4 4 ta ko‘rsatkichdan 2 tasida oldinda, RTX A5000 esa 1 tasida. L4 modelida FP32 (30,3 va 27,8 TFLOPS) hamda FP16/BF16 (121 va 111,1 TFLOPS) yuqoriroq. Biroq RTX A5000 modelida xotira o‘tkazuvchanligi yuqoriroq (0,77 va 0,3 TB/s). Bir xil: xotira hajmi.
- [NVIDIA L40S](https://uzdynamics.ai/uz/gpus/l40s)
  48 GB GDDR6 · 0,86 TB/s · PCIe 4
  L40S barcha 5 ta ko‘rsatkichda oldinda. L40S modelida FP8 (733 va 242 TFLOPS) hamda FP32 (91,6 va 30,3 TFLOPS) yuqoriroq.

Hisob: NVIDIA RTX A5000 uchun ko‘rsatgan siyraklik bilan FP16 qiymatining yarmi (222,2 TFLOPS); siyrakliksiz qiymatni NVIDIA keltirmaydi.

## Katalogdagi barcha GPU modellari

- [B300](https://uzdynamics.ai/uz/gpus/b300)
- [B200](https://uzdynamics.ai/uz/gpus/b200)
- [H200](https://uzdynamics.ai/uz/gpus/h200)
- [H100](https://uzdynamics.ai/uz/gpus/h100)
- [A100](https://uzdynamics.ai/uz/gpus/a100)
- [RTX PRO 6000](https://uzdynamics.ai/uz/gpus/rtx-pro-6000)
- [L40S](https://uzdynamics.ai/uz/gpus/l40s)
- [RTX 6000 Ada](https://uzdynamics.ai/uz/gpus/rtx-6000-ada)
- [RTX A6000](https://uzdynamics.ai/uz/gpus/rtx-a6000)
- [RTX A5000](https://uzdynamics.ai/uz/gpus/rtx-a5000)

[L4 uchun ariza](https://uzdynamics.ai/uz/request?source=gpu-page-end%3Al4&gpu=l4)

Vazifangizga mos ijara shartlarini yuboramiz. Shartnoma tuzilguncha to‘lov talab qilinmaydi.

```json
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"UZ Dynamics","item":"https://uzdynamics.ai/uz"},{"@type":"ListItem","position":2,"name":"GPU katalogi","item":"https://uzdynamics.ai/uz/gpus"},{"@type":"ListItem","position":3,"name":"L4","item":"https://uzdynamics.ai/uz/gpus/l4"}]}
```