Other · Microsoft
Microsoft Surface RTX Spark Dev Box
A small Windows 11 desktop built on NVIDIA's RTX Spark chip with 128GB of unified memory, on pre-order. It is the Windows counterpart to the GB10 boxes that ship with DGX OS.
Memory for models
The maker publishes no GPU limit for this coherent memory pool; 8 GB is assumed for the operating system.
Memory bandwidth
Maker price
Largest at 4-bit
113.1 GB at 8,192 tokens of context (an assumed limit, see Memory for models)
gpt-oss-20b, 4-bit
no bandwidth figure to estimate from
Memory configurations
One page covers every memory option the maker sells. Prices are the maker’s list price for the cheapest configuration with that much memory.
| Memory | Price | Memory the GPU can use |
|---|---|---|
| 128 GB | ~$6,000 | 120 GB gpu-usable memoryassumption |
The maker publishes no GPU limit for this coherent memory pool; 8 GB is assumed for the operating system. How GPU memory limits work on unified-memory machines.
What it runs at 4-bit
Every model in the catalog at its 4-bit file with 8,192 tokens of context. 30 of 41 fit. Speeds are the median of measured runs on the same chip where any exist; otherwise an estimate from memory bandwidth. Try other quants and context lengths.
| Model | Needs | Fit | Generation speed |
|---|---|---|---|
| Kimi K32.8T, 104B active · UD-Q4_K_XL | 1406.8 GBcontext memory estimated | Does not fit | |
| DeepSeek V4 Pro1.6T, 49B active · Q4_K_M | 890.5 GBcontext memory not counted | Does not fit | |
| Kimi K2.61T, 32B active · UD-Q4_K_XL | 545.7 GBcontext memory estimated | Does not fit | |
| DeepSeek V4.1 Flash552B, 16B active · Q4 (DwarfStar) | 484.5 GB | Does not fit | |
| GLM-5.3753.3B · UD-Q4_K_XL | 437.4 GBcontext memory estimated | Does not fit | |
| DeepSeek V3.2671B, 37B active · Q4_K_M | 379.6 GBcontext memory estimated | Does not fit | |
| MiniMax M3428B, 23B active · Q4_K_M | 244.8 GBcontext memory not counted | Does not fit | |
| Qwen3.5 397B-A17B397B, 17B active · Q4_K_M | 229.1 GBcontext memory estimated | Does not fit | |
| GLM-5.3 Flash320B, 18B active · UD-Q4_K_XL | 187.6 GBcontext memory estimated | Does not fit | |
| Hy3295B, 21B active · Q4_K_M | 173.6 GB | Does not fit | |
| MiniMax M2.7228.7B · Q4_K_M | 132.7 GB | Does not fit | |
| Qwen3.8 Flash-Next125B, 6B active · Q4_K_M | 113.1 GBcontext memory estimated | Fitsassumed limit | – |
| Nemotron 3 Super 120B-A12B120B, 12B active · Q4_K_M | 82.6 GBcontext memory estimated | Fitsassumed limit | – |
| Qwen3.5 122B-A10B122B, 10B active · Q4_K_M | 73.0 GBcontext memory estimated | Fitsassumed limit | – |
| Mistral Medium 3.5 128B128B · Q4_K_M | 74.0 GB | Fitsassumed limit | – |
| GLM-4.5-Air106B, 12B active · Q4_K_M | 70.9 GB | Fitsassumed limit | – |
| Mistral Small 4 119B (2603)119B, 6.5B active · Q4_K_M | 69.3 GBcontext memory estimated | Fitsassumed limit | – |
| Llama 4 Scout 17B-16E Instruct109B, 17B active · Q4_K_M | 62.4 GBcontext memory not counted | Fitsassumed limit | – |
| gpt-oss-120b117B, 5.1B active · MXFP4 (native) | 60.2 GBcontext memory estimated | Fitsassumed limit | – |
| Qwen3 Coder Next80B, 3B active · Q4_K_M | 46.9 GBcontext memory estimated | Fitsassumed limit | – |
| Llama 3.3 70B Instruct70B · Q4_K_M | 43.6 GB | Fitsassumed limit | – |
| Nemotron 3.5 Lightning 30B-A3B30B, 3B active · Q4_K_M | 25.3 GBcontext memory estimated | Fitsassumed limit | – |
| Qwen3.6 35B-A3B35B, 3B active · Q4_K_M | 22.4 GBcontext memory estimated | Fitsassumed limit | – |
| Ornith 1.5 35B-A3B35B, 3B active · Q4_K_M | 21.9 GBcontext memory estimated | Fitsassumed limit | – |
| Qwen3 Coder 30B-A3B Instruct30.5B, 3.3B active · Q4_K_M | 19.5 GB | Fitsassumed limit | – |
| Gemma 4 31B IT30.7B · Q4_K_M | 18.9 GBcontext memory estimated | Fitsassumed limit | – |
| Granite 4.2 30B29.3B · Q4_K_M | 20.3 GB | Fitsassumed limit | – |
| Qwen3.8 27B27B · Q4_K_M | 18.2 GBcontext memory estimated | Fitsassumed limit | – |
| Gemma 4 26B-A4B IT25.2B, 3.8B active · Q4_K_M | 17.4 GBcontext memory estimated | Fitsassumed limit | – |
| Qwen3.6 27B27B · Q4_K_M | 17.7 GBcontext memory estimated | Fitsassumed limit | – |
| Devstral Small 2 24B (2512)24B · Q4_K_M | 16.1 GB | Fitsassumed limit | – |
| gpt-oss-20b21B, 3.6B active · MXFP4 (native) | 12.5 GBcontext memory estimated | Fitsassumed limit | – |
| Phi-414.7B · Q4_K_M | 11.3 GB | Fitsassumed limit | – |
| Ministral 3 14B Reasoning (2512)13.9B · Q4_K_M | 10.4 GB | Fitsassumed limit | – |
| Gemma 4 12B IT11.95B · Q4_K_M | 8.2 GBcontext memory estimated | Fitsassumed limit | – |
| Qwen3.5 9B9B · Q4_K_M | 7.0 GBcontext memory estimated | Fitsassumed limit | – |
| Granite 4.2 8B8.8B · Q4_K_M | 7.9 GB | Fitsassumed limit | – |
| Gemma 4 E4B IT8B · Q4_K_M | 6.5 GBcontext memory not counted | Fitsassumed limit | – |
| Qwen3 8B8.2B · Q4_K_M | 7.3 GB | Fitsassumed limit | – |
| Llama 3.1 8B Instruct8B · Q4_K_M | 7.1 GB | Fitsassumed limit | – |
| Qwen3.5 4B4B · Q4_K_M | 4.3 GBcontext memory estimated | Fitsassumed limit | – |
Measured speeds
Nobody has published a run on this chip with a stated model, quant and backend that we could find. The speeds above are estimates.
Specifications
- Chip
- NVIDIA RTX Spark (Grace CPU + Blackwell RTX GPU)
- Memory
- unified (type not published)
- Memory bandwidth
- not published
- FP16 compute
- Microsoft quotes up to one petaflop of AI compute (FP4 with sparsity per its RTX Spark footnotes); no dense FP16 figure
- Software
- CUDA
- Ships with
- Windows 11 Pro
- Power
- not published
- Form factor
- Mini PC
- Networking
- Ethernet (speed not published); two USB-C
- Maker page
- microsoft.com
Sources
- microsoft.com: From $5,999.99 pre-order, 128GB unified memory, up to one petaflop AI compute, 100W thermal envelope, Windows 11 Pro, Ethernet
- microsoft.com: RTX Spark footnotes: TOPS figures are FP4 with sparsity; GPU-addressable maximum is less than total memory
Similar machines
- 2x Radeon RX 7900 XTX~$5,637
- Dell Pro Precision with GB10~$5,523
- Apple Mac Studio (M5 Ultra, 64-core GPU)~$5,499–$9,499