M2 Max 64GB vs NVIDIA RTX 5090 for local LLMs
20 reports on the M2 Max 64GB and 170 on the NVIDIA RTX 5090
Which is faster for local LLMs?
On Qwen3.6 35B · 3B active at 4-bit, the M2 Max 64GB has 77.5 generation tokens/s (2 runs); the NVIDIA RTX 5090 has 215.1 generation 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 4.4x higher on the NVIDIA RTX 5090 (410 GB/s on the M2 Max 64GB, 1,792 on the NVIDIA RTX 5090). Across 4 shared models with generation measurements on both cards, the M2 Max 64GB's median is higher on 0 and the NVIDIA RTX 5090's on 4.
- Memory
- 64GB unified
- Memory bandwidth
- 410 GB/s
- FP16 compute
- 13.6 TFLOPS
- Reports
- 20
- Memory
- 32GB
- Memory bandwidth
- 1,792 GB/s
- FP16 compute
- 419 TFLOPS
- Reports
- 170
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 | M2 Max 64GB | NVIDIA RTX 5090 |
|---|---|---|
| Qwen3.6 35B · 3B active4-bit | 77.52 runsLM Studio, Ollama32K to 256K context | 215.1one runllama.cppcontext not stated |
| Gemma 4 26B · 4B active4-bit | 61.6one runPP 465.2 (one run)MLXcontext not stated | 243.9one runPP 13809.2 (one run)llama.cppcontext not stated |
| Qwen3.5 27B4-bit | 20.7one runPP 91.4 (one run)MLXcontext not stated | 30.1one runPP 1515.8 (one run)llama.cpp1K context |
| Qwen3.5 35B · 3B active4-bit | 88.9one runPP 521.5 (one run)MLXcontext not stated | 194.0one runllama.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 | M2 Max 64GB | NVIDIA RTX 5090 |
|---|---|---|
| Qwen3.6 35B · 3B active | 32.9 t/sQ8_0 | 163.0 t/sQ5_K_M |
| Gemma 4 26B · 4B active | 28.2 t/sQ8_0 | 114.5 t/sQ6_K |
| Qwen3.5 27B | 10.4 t/sQ8_0 | 44.9 t/sQ6_K |
| Qwen3.5 35B · 3B active | 32.9 t/sQ8_0 | 163.0 t/sQ5_K_M |
| Qwen3.8 27B | 10.4 t/sQ8_0 | 44.9 t/sQ6_K |
| Qwen3.8 125B · 6B active | 47.5 t/sQ2_K | 20.3 t/sQ2_K, experts in RAM |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.6 27B | 10.4 t/sQ8_0 | 44.9 t/sQ6_K |
| DeepSeek V4.1 Flash 552B · 16B active | doesn't fit at these settings | doesn't fit at these settings |
| GLM-5.3 320B · 18B active | doesn't fit at these settings | doesn't fit at these settings |
Change the context, RAM or model in the calculator: M2 Max 64GB · NVIDIA RTX 5090
Every report on each card, with its source: M2 Max 64GB · NVIDIA RTX 5090