Engines for Apple M3 Ultra
16 engines run on Apple M3 Ultra.

Decodes faster and starts answering sooner on Apple Silicon, using an MTP drafter.

Runs large models such as DeepSeek V4 Flash and GLM 5.3 Flash on Apple Silicon, NVIDIA and AMD.

Loads GGUF models in pure Rust, with quantized Metal kernels and lossless speculative decoding.

Speeds up GLM-5.3-Flash on Apple M3 in a ds4 fork.
Fork of antirez/ds4
Serves as the reference upstream and runs on Apple Silicon, NVIDIA, AMD, Intel and CPU.

Runs models in Rust with ISQ and UQFF quantization on Apple Silicon and NVIDIA GPUs.

Serves as the reference upstream for Apple Silicon, built on MLX.

Runs on Apple Silicon and is the upstream of the MTPLX M1-family build.

Serves as the reference upstream: a UX wrapper with a model library.

Speeds up prompt reading on Apple Silicon by pairing a second Mac.

Offloads work to Core ML on Apple Silicon and to the XDNA NPU on AMD, in a llama.cpp fork.
Fork of ggml-org/llama.cpp
Builds on mlx-lm to decode Qwen3.5-9B faster than oMLX on Apple Silicon.
Built on ml-explore/mlx-lm
Decodes Qwen3.8-27B and Qwen3.6-35B-A3B faster than llama.cpp on Apple M3 and newer.

Serves as the reference upstream for Strata forks on NVIDIA, AMD and Apple Silicon.

Streams experts from the SSD to cut GPU memory use on Apple Silicon Macs.

Cuts memory use on Apple Silicon by loading only some of each MoE model's experts.