Qwen3.8-27B-MTPLX-Optimized-Speed-FP16

MTPLX: the fastest way to run Qwen 3.8 on a Mac. Native multi-token-prediction speculative decoding on Apple Silicon, two to three times the speed of plain decoding, exact at any temperature.

Qwen 3.8 27B Optimized Speed FP16

4-bit dynamic quant. Great coding speeds and good quality. Recommended. This is the M1 and M2 build.

The FP16 precision sibling of Qwen 3.8 27B Optimized Speed. M1 and M2 Macs do not run bf16 well, so this build keeps every quantized weight byte-identical to the parent and stores the remaining floating tensors (scales, biases, norms, the GDN convolution and state parameters, and the MTP head) in fp16 instead of bf16. Same layout, same tuned depth and draft settings, same context window. On an M3 or newer Mac use the parent instead.

MTPLX picks the right one for you: the app and mtplx start route M1 and M2 Macs to the FP16 builds and everything newer to the parents. It is the default MTPLX picks on an M1 or M2 Mac with 32 GB or more.

Measured on MTPLX 2.11.3 (16 September 2026)

This is a Qwen3.8-27B MLX pack for MTPLX, the fastest way to run Qwen 3.8 27B on a Mac. All numbers on a MacBook Pro M5 Max, fans verified at maximum, sampled at the model's own settings. Conditions and sources: mtplx.com/benchmarks.

Pack quality against the bf16 Qwen3.8-27B checkpoint, teacher-forced over 2,389 positions of code, prose, JSON and a multilingual notice (16 September 2026): Optimized Speed (4-bit dynamic) 96.0 percent top-1 agreement and KL 0.012; Optimized Quality (8-bit dynamic) 99.3 percent and KL 0.0005. Exactness on this release: a thousand four-token draws from the fast path match a thousand from the plain path within the plain path's own noise, at temperature 1, top-p 0.95, top-k 20. Details: MTPLX 2.11.3 release notes.

Runs on Apple Silicon Macs with 32 GB of unified memory or more (36 GB for Optimized Quality): MacBook Pro, MacBook Air, Mac mini and Mac Studio on M1 to M5. Guide: Run Qwen 3.8 27B on a Mac. Comparison pages: MTPLX vs mlx-serve, MTPLX vs oMLX, MTPLX vs LM Studio, MTPLX vs Ollama.

Speeds

The numbers we publish for the parent were measured on an M5 Max: 58.7 tok/s on the coding task and 35.1 to 37.3 tok/s on long xhigh reasoning, official Qwen 3.8 sampling, generation running to the model's own stop. This FP16 build has the same weights and runs the same MTPLX turbo path, so the speculative math is identical; absolute tok/s on an M1 or M2 depends on that chip. We have not published M1 or M2 numbers for it yet.

MTPLX_FP16_CONVERSION_MANIFEST.json in the repo lists every tensor that was cast and every tensor that was preserved, with sha256 for each shard. Speculation in MTPLX is exact at any temperature: drafts are accepted with the probability-ratio rule plus residual resampling.

Use it

Mac app: download at mtplx.com, pick "Qwen 3.8 27B Optimized Speed FP16".

Command line:

pip install mtplx
mtplx serve --model Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed-FP16

The other two FP16 builds: Bare Speed FP16, Optimized Quality FP16.