AMD Gorgon Halo 192GB vs M2 Ultra 192GB for local LLMs
6 reports on the AMD Gorgon Halo 192GB and 8 on the M2 Ultra 192GB
Which is faster for local LLMs?
On MiniMax M2.7 230B · 10B active at 4-bit, the AMD Gorgon Halo 192GB has 30.2 generation tokens/s (one run); prompt processing 309.0 tokens/s (one run); the M2 Ultra 192GB has 32.7 generation tokens/s (one run); prompt processing 420.8 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 3.0x higher on the M2 Ultra 192GB (273 GB/s on the AMD Gorgon Halo 192GB, 819 on the M2 Ultra 192GB).
- Memory
- 192GB unified
- Memory bandwidth
- 273 GB/s
- FP16 compute
- 61.4 TFLOPS
- Reports
- 6
- Memory
- 192GB unified
- Memory bandwidth
- 819 GB/s
- FP16 compute
- 27.2 TFLOPS
- Reports
- 8
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 | AMD Gorgon Halo 192GB | M2 Ultra 192GB |
|---|---|---|
| MiniMax M2.7 230B · 10B active4-bit | 30.2one runPP 309.0 (one run)llama.cppcontext not stated | 32.7one runPP 420.8 (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 | AMD Gorgon Halo 192GB | M2 Ultra 192GB |
|---|---|---|
| MiniMax M2.7 230B · 10B active | 8.2 t/sQ5_K_S | 37.1 t/sQ3_K_M |
| Qwen3.8 27B | 2.8 t/sFP16 | 8.3 t/sFP16 |
| Qwen3.8 125B · 6B active | 9.5 t/sQ8_0 | 33.0 t/sQ6_K |
| DeepSeek V4 Flash 284B · 13B active | 12.5 t/sQ4_K_M | 62.8 t/sQ3_K_S |
| Qwen3.6 35B · 3B active | 10.0 t/sFP16 | 29.9 t/sFP16 |
| Qwen3.6 27B | 2.8 t/sFP16 | 8.3 t/sFP16 |
| Gemma 4 26B · 4B active | 7.5 t/sFP16 | 22.5 t/sFP16 |
Change the context, RAM or model in the calculator: AMD Gorgon Halo 192GB · M2 Ultra 192GB
Every report on each card, with its source: AMD Gorgon Halo 192GB · M2 Ultra 192GB