M3 Ultra 512GB vs AMD Radeon AI PRO R9700 32GB for local LLMs
41 reports on the M3 Ultra 512GB and 63 on the AMD Radeon AI PRO R9700 32GB
Which is faster for local LLMs?
On Qwen3.8 27B at 4-bit, the M3 Ultra 512GB has 31.8 generation tokens/s (one run); prompt processing 391.2 tokens/s (one run); the AMD Radeon AI PRO R9700 32GB has 29.8 generation tokens/s (3 runs); prompt processing 1287.0 tokens/s (one run). Those are community reports whose engines and settings differ, so the gap is a rough sign, not a controlled measure of the two cards. Memory bandwidth, which caps how fast a card can write, is 28% higher on the M3 Ultra 512GB (819 GB/s on the M3 Ultra 512GB, 640 on the AMD Radeon AI PRO R9700 32GB). Across 2 shared models with generation measurements on both cards, the M3 Ultra 512GB's median is higher on 1 and the AMD Radeon AI PRO R9700 32GB's on 1.
- Memory
- 512GB unified
- Memory bandwidth
- 819 GB/s
- FP16 compute
- 28.4 TFLOPS
- Reports
- 41
- Memory
- 32GB
- Memory bandwidth
- 640 GB/s
- FP16 compute
- 191 TFLOPS
- Reports
- 63
Measured on both
Generation and prompt processing speeds from plain runs (one device, one request, no speculative decoding, the whole model in the device's memory) of the same model size at the same bit class. Each side shows the engines and reported context behind its median. Context can mean a configured window or prompt depth, so these are community comparisons with differing settings. How to read these
| Model | M3 Ultra 512GB | AMD Radeon AI PRO R9700 32GB |
|---|---|---|
| Qwen3.8 27B4-bit | 31.8one runPP 391.2 (one run)MLXcontext not stated | 29.83 runsPP 1287.0 (one run)llama.cpp, 1 not stated2K context, 2 not stated |
| Qwen3.5 35B · 3B active4-bit | 80.9one runPP 2779.0 (one run)llama.cppcontext not stated | 149.2one runPP 3075.0 (one run)llama.cppcontext not stated |
| Gemma 4 26B · 4B active8-bit | 84.9one runPP 2634.8 (one run)llama.cppcontext not stated | -0 runsPP 3200.0 (one run)llama.cppcontext not stated |
Estimated by the calculator
The quant the calculator recommends for each card and its estimated speed at a 32,768-token context with 32 GB of system RAM, from memory bandwidth and the model's shape. Where a model was also measured above, the measurement wins.
| Model | M3 Ultra 512GB | AMD Radeon AI PRO R9700 32GB |
|---|---|---|
| Qwen3.8 27B | 8.0 t/sFP16 | 16.0 t/sQ6_K |
| Qwen3.5 35B · 3B active | 28.7 t/sFP16 | 58.2 t/sQ5_K_M |
| Gemma 4 26B · 4B active | 21.5 t/sFP16 | 40.9 t/sQ6_K |
| Qwen3.8 125B · 6B active | 13.6 t/sFP16 | 18.7 t/sQ2_K, experts in RAM |
| Qwen3.6 35B · 3B active | 28.7 t/sFP16 | 58.2 t/sQ5_K_M |
| DeepSeek V4 Flash 284B · 13B active | 22.6 t/sQ8_0 | doesn't fit at these settings |
| Qwen3.6 27B | 8.0 t/sFP16 | 16.0 t/sQ6_K |
| DeepSeek V4.1 Flash 552B · 16B active | 30.0 t/sQ4_K_M | doesn't fit at these settings |
| GLM-5.3 320B · 18B active | 7.4 t/sQ8_0 | doesn't fit at these settings |
Change the context, RAM or model in the calculator: M3 Ultra 512GB · AMD Radeon AI PRO R9700 32GB
Every report on each card, with its source: M3 Ultra 512GB · AMD Radeon AI PRO R9700 32GB