---
title: "O‘zbekistonda B200 ijarasi: dastlabki kirish | UZ Dynamics"
description: "NVIDIA B200: ariza orqali dastlabki kirish, mavjudlik va form-faktor aniqlashtirilmoqda. 180 GB HBM3e · 7,7 TB/s · NVLink."
url: "https://uzdynamics.ai/uz/gpus/b200"
lang: "uz"
image: "https://uzdynamics.ai/uz/gpus/b200/opengraph-image/card?f08c4f789845417d"
alternates:
  ru: "https://uzdynamics.ai/ru/gpus/b200"
  uz: "https://uzdynamics.ai/uz/gpus/b200"
  en: "https://uzdynamics.ai/en/gpus/b200"
---

> 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 B200 ijarasi: dastlabki kirish

Blackwell · SXM (aniqlanmoqda)

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

B200

[B200 uchun ariza](https://uzdynamics.ai/uz/request?source=gpu-page%3Ab200&gpu=b200)

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

- **Har bir GPU xotirasi:** 180 GB HBM3e
- **Xotira o‘tkazuvchanligi:** 7,7 TB/s
- **Kartalar orasidagi aloqa:** NVLink 1,8 TB/s

B200 sxemasi: 180 GB HBM3e, chipning eng katta hajmi — 192 GB; NVLink 1,8 TB/s

Soddalashtirilgan sxema, masshtabga rioya qilinmagan

## NVIDIA B200 xususiyatlari

- **Arxitektura:** Blackwell
- **E’lon qilingan yil:** 2024
- **Xotira:** 180 GB NVIDIA ma’lumotlariga ko‘ra foydalanish mumkin bo‘lgan xotira. Chip xotirasining eng katta hajmi — 192 GB.
- **Xotira turi:** HBM3e
- **Xotira o‘tkazuvchanligi:** 7,7 TB/s NVIDIA Blackwell Technical Brief hujjati bo‘yicha (HGX B200); NVIDIA’ning DGX B200 materiallarida — 8 TB/s gacha.
- **FP32:** 75 TFLOPS
- **FP16/BF16 siyrakliksiz:** 2 250 TFLOPS
- **FP8 siyrakliksiz:** 4 500 TFLOPS
- **FP4 siyrakliksiz:** 9 000 TFLOPS
- **FP64:** 37 TFLOPS
- **Tenzor yadrolarida FP64:** 37 TFLOPS
- **RT yadrolari:** yo‘q
- **NVENC/NVDEC:** NVENC 0, NVDEC 7
- **MIG:** ko‘pi bilan 7 ta MIG bo‘linmasi
- **Kartalar orasidagi aloqa:** NVLink 1,8 TB/s
- **Quvvat (TDP/TGP):** 1 000 Vt gacha
- **Form-faktor:** SXM (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.

B200 form-faktori aniqlashtirilmoqda. Xususiyatlar NVIDIA rasmiy materiallari bo‘yicha, 26-sentabr, 2026 sanasida tekshirilgan.

## B200 vazifalar bo‘yicha

- Inferens: 123 mlrd parametrgacha LLM (FP8), bitta kartada
- Qo‘shimcha o‘qitish: QLoRA 235 mlrd parametrgacha, LoRA 32 mlrd parametrgacha, bitta kartada
- Noldan o‘qitish: 8 mlrd parametrgacha model, bitta kartada
- Render: RT yadrolari yo‘q
- FP64 hisob-kitoblari: 37 TFLOPS

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

## NVIDIA B200 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%3Ab200&gpu=b200&count=more)

## Manbalar

- [NVIDIA Blackwell Technical Brief](https://dam-cdn.nvd.orangelogic.com/AssetLink/gl2l4l4812s5fw0p614s6i8bv6mi3vx5.pdf)
- [NVIDIA HGX Platform](https://www.nvidia.com/en-us/data-center/hgx/)
- [NVIDIA DGX B200](https://www.nvidia.com/en-us/data-center/dgx-b200/)
- [NVIDIA Multi-Instance GPU User Guide](https://docs.nvidia.com/datacenter/tesla/mig-user-guide/latest/supported-gpus.html)
- [NVIDIA Video Encode and Decode Support Matrix](https://developer.nvidia.com/video-encode-decode-support-matrix)
- [NVIDIA Omniverse Developer Guide: Technical Requirements](https://docs.omniverse.nvidia.com/dev-guide/latest/common/technical-requirements.html)
- [NVIDIA Newsroom, March 18, 2024](https://nvidianews.nvidia.com/news/nvidia-blackwell-platform-arrives-to-power-a-new-era-of-computing)

## O‘xshash GPU modellari

- [NVIDIA H200](https://uzdynamics.ai/uz/gpus/h200)
  141 GB HBM3e · 4,8 TB/s · NVLink
  B200 7 ta ko‘rsatkichdan 6 tasida oldinda, H200 esa 1 tasida. B200 modelida FP16/BF16 (2 250 va 989,4 TFLOPS) hamda FP8 (4 500 va 1 978,9 TFLOPS) yuqoriroq. Biroq H200 modelida tenzor yadrolaridagi FP64 unumdorligi yuqoriroq (67 va 37 TFLOPS). Tenzor yadrolarisiz FP64 unumdorligi B200 modelida yuqoriroq: 37 va 34 TFLOPS.
- [NVIDIA B300](https://uzdynamics.ai/uz/gpus/b300)
  270 GB HBM3e · 7,7 TB/s · NVLink
  B300 8 ta ko‘rsatkichdan 2 tasida oldinda, B200 esa 1 tasida. B300 modelida FP4 (14 000 va 9 000 TFLOPS) hamda xotira hajmi (270 va 180 GB) yuqoriroq. Biroq B200 modelida FP64 yuqoriroq (37 va 1,2 TFLOPS). Bir xil: xotira o‘tkazuvchanligi, FP32, FP16/BF16, FP8, NVLink.
- [NVIDIA RTX PRO 6000 Blackwell](https://uzdynamics.ai/uz/gpus/rtx-pro-6000)
  96 GB GDDR7 · 1,6 TB/s · PCIe 5
  B200 6 ta ko‘rsatkichdan 5 tasida oldinda, RTX PRO 6000 esa 1 tasida. B200 modelida xotira o‘tkazuvchanligi (7,7 va 1,6 TB/s) hamda FP16/BF16 (2 250 va 467,8 TFLOPS) yuqoriroq. Biroq RTX PRO 6000 modelida FP32 yuqoriroq (120 va 75 TFLOPS). FP64: B200 modelida 37 TFLOPS, RTX PRO 6000 uchun NVIDIA karta spetsifikatsiyasida ko‘rsatilmagan (NVIDIA texnik hujjatiga ko‘ra GB202 chipida FP32 ning 1/64 qismi).

Siyrakliksiz FP16/BF16 va FP8 — NVIDIA texnik hujjatidagi H100 SXM5 ustunidan: H200 uchun NVIDIA faqat siyraklik bilan olingan qiymatlarni beradi, ular H100 SXM qiymatlari bilan bir xil.

RTX PRO 6000 uchun FP16, FP8 va FP4 — hisoblangan qiymatlar. Hisob: NVIDIA texnik hujjatidagi har bir SM uchun tenzor yadrolari unumdorligi × mahsulot tavsifidagi 2 430 MHz Boost chastotasi. Server Edition uchun NVIDIA FP16/FP8/FP4 ko‘rsatkichlarini faqat yaxlitlangan 1/2/4 PFLOPS maksimal qiymatlari sifatida keltiradi. Bu siyraklikdan foydalanilgandagi qiymatlar darajasiga mos; siyrakliksiz qiymatlar e’lon qilinmagan.

## Katalogdagi barcha GPU modellari

- [B300](https://uzdynamics.ai/uz/gpus/b300)
- [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)
- [L4](https://uzdynamics.ai/uz/gpus/l4)
- [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)

[B200 uchun ariza](https://uzdynamics.ai/uz/request?source=gpu-page-end%3Ab200&gpu=b200)

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":"B200","item":"https://uzdynamics.ai/uz/gpus/b200"}]}
```