Назад
HARDWARE / UNIFIED MEMORY

NVIDIA DGX Spark · 128 GB

A compact local AI system with 128 GB coherent memory. It can hold large low-bit models, but decode speed and software support remain model-specific.

Первичный источник
GB10 Grace Blackwell128 GB
01 / PRACTICAL CEILING

Qwen3.8 Flash Next REAP-288

The highest listed result may be a tight or offloaded fit. For daily use, prefer the first result marked Comfortable and keep context modest.

02 / DEFAULT RUNTIME

vLLM

Linux GPU servers, concurrency and OpenAI-compatible production APIs. CUDA, ROCm and XPU feature coverage differs. Driver, shared-memory and quantization compatibility are version-specific.

Открыть гид
03 / MODEL ENVELOPE

Показательные модели с открытыми весами

Community REAP build

Qwen3.8 Flash Next REAP-288

176BREAP-288 · MLX 4-bit · PLE on NVMe
Оценка памяти
~40.57 GB
Контекст
7.6K recommended on measured 48 GB Mac
OpenAI

gpt-oss-120b

117B5.1B active · MXFP4
Оценка памяти
~80 GB
Контекст
128K
Meta

Llama 4 Scout

109B17B active · INT4 MoE
Оценка памяти
~78 GB
Контекст
10M advertised
Meta

Llama 3.1 70B

70BGGUF Q4
Оценка памяти
~46 GB
Контекст
128K
Alibaba Qwen

Qwen3.6 35B-A3B

35B3B active · Q4 / NVFP4 MoE
Оценка памяти
~25 GB
Контекст
262K native
Alibaba Qwen

Qwen3 32B

32BGGUF Q4
Оценка памяти
~24.5 GB
Контекст
32K+
Внимание

Fit is based on estimated total model memory. Driver support, KV cache, multimodal projectors, concurrency, and desktop applications can all reduce available headroom. Current fit: Qwen3.8 Flash Next REAP-288 (Comfortable), gpt-oss-120b (Comfortable), Llama 4 Scout (Comfortable), Llama 3.1 70B (Comfortable), Qwen3.6 35B-A3B (Comfortable), Qwen3 32B (Comfortable).