llamaperf

Mac M6 reports

23 reports. These are individual setups, not a controlled benchmark. M6 configurations and speeds →

Speeds are as each source reported them. GPU count, offloading and concurrent requests change them, and prompt-processing speed is input throughput, not time to first token. Check the full setup before comparing two reports.

Apple hardware
generation:
16.2 tokens/s
prompt processing (prefill):
439.0 tokens/s
quant:
affine Q4, group size 64 (MLX)

User reports Qwen3.5-4B at 16.24 t/s generation and 438.97 t/s prefill on an Apple M6 Mac mini with 32GB unified memory. Setup is MLX 0.32.2 with MLX-LM 0.31.3, a controlled affine 4-bit (Q4) MLX package with group size 64 and 16-bit KV cache, single request, 2048 input tokens and 128 generated tokens. Measurements are from Precisit's published benchmark, not an independent rerun; decode excludes first-token production and model loading, and the 2048-token input is not a configured context limit.

Oct 8, 2026
Apple hardware
generation:
52.2 tokens/s
quant:
4-bit (MLX)

User reports Gemma-4-26B-A4B 4-bit at 52.2 tok/s decode on a base Mac mini M6 with 32 GB unified memory. Setup is SwiftLM with MLX 4-bit weights, full GPU offload, peak GPU memory 19.5 GB; the longest prompt that passed was 80.7K tokens at 622 tok/s prefill and 24.3 tok/s decode. Also benchmarks Qwen3.6-35B-A3B 4-bit at 46.7 tok/s (GPU) and 13.2 tok/s (--stream-experts), Qwen3.8-27B 4-bit dense at 9.3 tok/s with 200 tok/s prefill, and Gemma-4-26B-A4B 8-bit at 8.8 tok/s with --stream-experts. An A/B against the prior revision shows 998 tok/s prefill at ~2.3K tokens versus 508 tok/s with swap, and 914 tok/s at ~9.5K tokens where the earlier build aborted on swap.

Oct 5, 2026

Qwen3.6 35B (3B active)

M6 32GB · LM Studio · 32,768 ctx

Tone: positiveApple hardware
generation:
63.8 tokens/s
quant:
MLX 4-bit (MLX)
summarizationmultilingualcodingtool-use

User reports Qwen3.6-35B-A3B at 63.8 t/s median decode on a Mac mini M6 with 32 GB unified memory. Setup is LM Studio with the MLX 4-bit build and a 32,768-token context; the MLX build ignored the context setting and loaded 34k to 165k. Reasoning off, max_tokens 600, temperature 0.7, median of 3 runs per prompt. The same model's Splash build (speculative decoding with DFlash2 drafts) reached 51.5-234.1 t/s across the five prompts, slower on German prose and faster on code and JSON. The user recommends Splash where it exists and notes the MLX build was faster than Splash on German prose (63 vs 52-59 t/s).

Oct 4, 2026