oMLX on the M5 Ultra 256GB
On the M5 Ultra 256GB, oMLX runs DeepSeek V4.1 Flash 552B · 16B active at 3-bit at a median 73.4 tokens per second for generation, from 1 plain run (context not stated).
Speed by model on the M5 Ultra 256GB
Plain runs only: one device, one request, no speculative decoding, the whole model in memory. Each row is one model size at one bit class, with the other engines' plain runs of the same model on this card beside it. Context length and build still differ between runs, and each side shows its own.
| Model | oMLX | Other engines |
|---|---|---|
| DeepSeek V4.1 Flash 552B · 16B active 3-bit | 73.4 t/sone report, context not stated | no other engine reported |
| Qwen3.8 27B 4-bit | 50.0 t/sone report, 8K context | no other engine reported |
7 oMLX reports on the M5 Ultra 256GB
Newest first. A run on several cards, with speculative decoding, with part of the model in system RAM or with batched requests says so, since each describes a different setup.
- Qwen3.8 125B · 6B active · oQ6e135.2 t/s(speculative decoding)
- GLM-5.3 320B · 18B active · oQ476.7 t/s(speculative decoding)
- GLM-5.3 320B · 18B active · oQ4e68.8 t/s(speculative decoding)
- DeepSeek V4.1 Flash 552B · 16B active · Q3-G128-LSQ73.4 t/s
- Qwen3.8 125B · 6B active · oQ8e29.0 t/s
- Qwen3.8 125B · 6B active · oQ4e108.0 t/s(speculative decoding)
- Qwen3.8 27B · q450.0 t/s
Frequently asked
How fast is oMLX on the M5 Ultra 256GB?
On the M5 Ultra 256GB, oMLX runs DeepSeek V4.1 Flash 552B · 16B active at 3-bit at a median 73.4 tokens per second for generation, from 1 plain run (context not stated).