Single Intel Arc Pro B70 with vLLM XPU cookbook
| GPU(s) | Intel Arc Pro B70 32 GB Amazon · eBay (used) · Newegg |
|---|---|
| VRAM / unified memory (GB) | 32 |
| CPU | |
| Motherboard | |
| RAM | ~12 GB free host RAM needed (embeddings in RAM) Amazon · eBay (used) · Newegg |
| PSU | |
| Other | vLLM XPU with B70 patches (ghcr.io/riceharvest/b70-vllm) |
| Reported total cost | |
| Power | Stock 275 W cap |
Measured speed (generation tokens/sec)
📏 How measured: reported by the original source; RigBase re-checked the numbers at that source
| Model | Quant | Engine | tok/s | Details |
|---|---|---|---|---|
| Qwen3.8 27B | GPTQ Int4 (asym) | vLLM XPU b70 image + MTP, reasoning prompt | 96 | ⚡ spec-decode (MTP)context: short prompt📏 Greedy, one request at a time, 512-token answers (README "Numbers" table) |
| Qwen3.8 27B | GPTQ Int4 (asym) | vLLM XPU b70 image + MTP, prose prompt | 66 | ⚡ spec-decode (MTP)context: short prompt📏 Greedy, one request at a time, 512-token answers (README "Numbers" table) |
| Qwen3.6 35B-A3B | GPTQ Int4 | vLLM XPU b70 image, no MTP | 120 | context: short prompt📏 Greedy, one request at a time, 512-token answers (README "Numbers" table) |
Greedy, one request at a time, 512-token answers, short prompt (README 'Numbers' table). Code prompts reach 113 tok/s; at 32K context ~72 tok/s. Qwen3.6 35B-A3B figure is the README's '~120 tok/s without MTP'. Found via localmaxxing (discovery only); numbers re-checked at the original source on 2026-10-24.
Source: https://github.com/riceharvest/b70-qwen38-cookbook
Blank fields = not stated in the source. We never estimate build data; GPU street prices above are separately sourced and dated.
Rent before you buy: try this model on a cloud GPU for a few dollars first: Runpod (referral link; sign up with Google to get the new-user bonus)
Quick answers: Can Intel Arc Pro B70 run Qwen3.6 35B-A3B? · Can Intel Arc Pro B70 run Qwen3.8 27B?