Qwen3.8 27B on a Mac Studio: 14 tok/s, but faster per answer
Run Qwen3.8 27B locally: real numbers from my Mac Studio
A Mac Studio M3 Ultra user benchmarks Qwen3.8 27B (Q4_K_M) via Ollama, finding 14 tok/s generation—half the speed of its predecessor—but the model uses a third of the tokens, making wall-clock time per answer a tie. The 1-bit quant runs at 27 tok/s but can't commit to answers. Includes RAM requirements per quant, hardware recommendations, and a llama.cpp update gotcha.
Slower per token, faster per answer.
- Youden
I've been thinking about buying a system to run LLMs locally but the price for one that'll run Qwen3.8-27B well is quite offputting to say the least.
What I've been looking at instead is inference providers that use TEE and E2EE to provide cryptographic guarantees that my prompts and responses are only visible to me and the GPU itself.
Despite their docs and assurances of what their guarantees mean, I'm having trouble getting to a point where I'm actually comfortable trusting them with secrets though. Phala for example seems to be E2EE only to the gateway and will then forward prompts to (potentially third party) providers.
Has anyone been down this path and found a provider they feel safe with?
- Infernal
I am really surprised you’re seeing half the generation performance of 3.6 with 3.8 at the same parameter size and quantization (and same prefill performance to boot) - is there just an optimization in the stack somewhere for 3.6 that hasn’t landed yet for 3.8?
Curious if other folks are also surprised by this apparent discrepancy or have a ready explanation.
- Atreiden
I'm encountering the same behavior. I've tried 4-8bit quants and get 14-17 tok/s with one run that achieved 19. I'm eagerly awaiting dflash2 support in Unsloth or LM Studio, as allegedly that should increase throughput to around 30tok/s, which is the baseline for what I consider at least somewhat interactive.
Jealous of the folks with 5090s running ninfer and getting >100tok/s. At those speeds it's a true frontier replacement IMO.
- rbanffy
Pretty impressive how the Mac ends up less expensive than the Strix Halo boxes, at least here in Ireland. A 128GB Mac Studio with an M5 Max (the Ultra can only have 96 or 256GB) still costs less than the "GMKtec EVO-X2" or the Nvidia DGX Spark with similar performance. Is it the same in the US?
- pwython
Yea Qwen3.8 wasn't fun to use on my 64GB M4 Max either (better than these numbers though), so my new daily driver is Ornith-1.5-35B-A3B-MLX-4bit. I recommend giving that a whirl if you're on similar hardware, it's definitely better than Qwen3.6 35b-a3b which was my go-to before.